筆者在論文中引入一個名為加權距離檢驗的雙變項統計。此檢驗方法用於比較兩個隨機變數的集中趨勢。其優勢在於在有偏度的數據中，仍能穩定地控制第一型錯誤，並同時提供可觀的統計檢力。加權距離檢驗利用冪函數修正在偏度數據中的不對稱現象。與一般的冪函數轉換法不同，加權距離檢驗將冪值限制在0和1之間。文中亦提供了一個有效決定冪值的方法，以方便在實際運算中使用。 / 筆者總結了四個主流的雙變項統計方法，並利用蒙地卡羅模擬法在正態分佈、同程度偏度分佈以及不同程度偏度分佈三個情況中比較了它們與加權距離檢驗的表現。結果顯示，加權距離檢驗雖然沒有在任何一個情況中勝出，但卻於兩方面表現了其優勢。首先，它在任何情況下都能把第一型錯誤控制在合理水平之下；其次，它在任何情況下都不至於表現得太差。反觀其他四個檢驗方法總會在某些情況下表現失敗。由此可見，加權距離檢驗比起其他檢驗方法更能提供一個穩定而簡單的方法去比較集中趨勢。 / A new bivariant statistics, namely the weighted distance test, for comparing two groups were introduced. The test aims at providing reliable type I error control and reasonable statistical power across different types of skewed data. It corrects the skewness of the data by applying power transformation with power index ranged between 0 to 1. I also proposed in this thesis a possible way of deciding the power index by considering the skewness difference between the two groups under comparison. / I reviewed 4 commonly used inferential statistics for two-group comparison and compared their performances with the weighted distance test under 1) normal distribution, 2) skewed distribution with equal skewness across groups, and 3) skewed distribution with unequal skewness across groups. Monte Carlo simulations were ran to evaluate the 5 tests. Results showed that the weighted distance test was not the best test in any particular situation, but was the most stable test in the sense that 1) it provided accurate type I error control and 2) it did not produce catastrophically small power in any scenario. All other 4 tests failed in some of the simulated scenario for either inflated type I error, or unsatisfactory power. Therefore, I suggested that the weighted distance test could be one easy-to-use test that works fairly well across a wide range of situation. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Lee, Yung Ho. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2012. / Includes bibliographical references (leaves 31-33). / Abstracts also in Chinese. / Chapter Chapter One --- : Introduction --- p.1 / Common methods in comparing central tendency --- p.2 / T-test --- p.2 / Median and rank --- p.3 / Trimming --- p.3 / Power transformation --- p.4 / Chapter Chapter Two --- : Weighted distance statistic --- p.5 / Definition --- p.5 / Statistical properties --- p.5 / Specification of Lambda λ --- p.7 / Estimation and inference --- p.7 / Chapter Chapter Three --- : Simulation --- p.10 / Study 1 --- p.12 / Study 2 --- p.15 / Study 3 --- p.18 / Chapter Chapter Four --- : Discussion --- p.21 / Summary --- p.21 / Limitation --- p.22 / Further development --- p.23 / Chapter Appendix I --- : Proofs of theorems of weighted distance statistic --- p.24 / Chapter Appendix II --- : Table of numerical results of simulations --- p.26 / Bibliography --- p.30

Many large-scale national and international testing programs use the Rasch model to govern the construction of measurement scales that can be used to monitor standards of performance and monitor performance over time. A significant issue that arises in such programs is that once a decision has been made to use the model, it is not possible to reverse the decision if the data do not fit the model. There are two levels of question that result from such a situation. One of them involves the issue of misfit to the model. That is, how robust is the model to violations of fit of the data to the model? A second question emerges from the premise that the issue of fit to the model is a relative matter. That is, ultimately, it becomes the decision of users as to whether data fit the model well enough to suit the purpose of the users. Once this decision has been made, such as in the case of large-scale testing programs like the ones refocused to above, then the question reverts to one in which the focus is on the applications of the Rasch model. More specifically, in the case of this study, the intention is to examine the consequences of variability of fit to the Rasch model on the measures of student performance obtained from two different equating procedures. Two related simulation studies have been conducted to compare the results obtained from using two different equating procedures (namely separate and concurrent equating) with the Rasch Simple Logistic model, as data-model fit gets progressively worse. The results indicate that when data-model fit ranges from good fit to average fit (MNSQ ??? 1.60), there is little or no difference between the results obtained from the different equating procedures. However, when data-model fit ranges from relatively poor fit to poor fit (MNSQ &gt 1.60), the results from using different equating procedures prove less comparable. When the results of these two simulation studies are translated to a situation in Australia, for example, where different states use different equating procedures to generate a single comparable score and then these scores are used to compare performances amongst students and to predetermined standards or benchmarks, it raises significant equity issues. In essence, it means that in the latter situation, some students are deemed to be either above or below the standards purely as a consequence of the equating procedure selected. For example, students could be deemed to be above a benchmark if separate equating was used to produce the scale; yet these same students could be deemed to fall below the benchmark if concurrent equating is used. The actual consequences of this decision will vary from situation to situation. For example, if the same equating procedure was used each year to equate the data to form a single scale, then it could be argued that it does not matter if the results vary from occasion to occasion because it is consistent for the cohort of students from year to year. However, if other states or countries, for example, use a different equating procedure and the results are compared, then there is an equity problem. The extent of the problem is dependent upon the robustness of the model to varying degrees of misfit.

The purpose of this study was to investigate the willingness of people to consider nontraditional sex roles, and to determine whether males and females were more prone to consider nontraditional sex roles when the person attempting to influence them was male or female. It was hypothesized that individuals can be influenced to consider nontraditional sex roles, and that male presenters would have greater impact than female presenters. It was also hypothesized that male presenters would be perceived as more credible and more adequate.The subjects were 93 undergraduate students, 42 males and 51 females, who volunteered to participate in the study, and were told that they would be "helping us to evaluate what and who should be included in a presentation for a national convention as well as a seminar on campus." They were divided into two groups on the basis of sex and then were randomly assigned to one of three conditions: one of two female speakers, one of two male speakers, and the control or music group.Subjects were given written explanations of the "project'' and written background information about the audiotapes to which they listened. Treatment groups listened to a tape containing either a male or a female advocating nontraditional sex roles while control subjects listened to a tape of music. After listening to the tapes, the treatment subjects responded to two semantic differential instruments (Quality of Presentation, and Person), and a generalized reaction questionnaire (which was essentially part of the deception). They then responded (by raising their hands to indicate yes or no) to three verbal questions concerning their commitment to their attitudes. They were then administered the final questionnaire, the Attitudes Toward Women Scale (AWS).After listening to the tape of music, the control subjects completed one semantic differential instrument (Quality of Performance), a generalized reaction questionnaire, and the AWS. They were given no verbal questions. All of the subjects were immediately debriefed together.Three analyses of variance and a series of univariate contrasts were performed on the primary dependent measure, the AWS, with significance being sought at the .05 level. Results indicated that the treatment was not effective. However, female subjects scored significantly higher (more liberal) than male subjects (F = 23.77, p < .001), and the male presenters as a group were more effective than the female presenters (F = 4.18, p < .05), and by extrapolation, more effective than the control condition.Multivariate analyses of variance and multivariate contrasts were performed on the secondary dependent measures, the semantic differential instruments (which were equally divided into three factors: activity, evaluation, and potency). There were no significant differences in the ratings given to male and female presenters. There was, however, a significant difference between the two female presenters on the activity factor (F = 8.49, p < .05). In addition, it appears that female subjects rated the presenters significantly higher (F = 7.29, p G .05) on the activity factor than did male subjects, and by contrast, male subjects gave higher ratings (F = 6.91, p < .05) on the evaluation factor than did female subjects.Two secondary analyses were performed on supplemental data that had been collected. The first analysis indicated that subjects' scores on the AWS were significantly correlated (r = -.1828, p < .05) with how close they said the presenters' views were to their own. The closer they perceived the views to be, the higher their AWS score. In the second analysis the subjects' answersto the three public commitment questions were compared with (1) the sex of the subjects, and (2) the individual presenters. There was a significant difference in the subsects' desires to meet the various speakers (X2 = 11.412, p < .05), emphasizing earlier trends indicating differences among the individual presenters.The recommendation was made for additional research in the area of influencing people to consider more flexible, nontraditional sex roles, and how responses are affected by the gender of the people involved. Investigations using a larger number of presenters, and more impactfully devised modes of treatment were suggested.

No description available.

No description available.

The present study sought to investigate the relationship between psychological reactance, as a component of paradoxical intention and locus of control.The Rotter I-E Scale, which measures individuals locus of control on a scale ranging from zero to 23, was utilized to determine subjects' external or internal ranking. A research design was developed which included a pilot study and an experiment which contained pretest and posttest conditions for examining subjects' choice change, From the pilot study two pictures were selected which were not biased towards internals or externals in the experimental conditions. Choice change was measured by subject's pretest choice against subject's posttest choice of the two pictures.The study was designed to answer the following research questions:1) Would there be a significant difference in the choices made by the experimental subjects from those made by the control subjects, thereby indicating that psychological reactance had occurred with the experimental group? 2) Would the findings of psychological the experimental reactance indicated in question one above, be significant if differences were compared by quartile scores on the I-E Scale?In the experiment, 122 experimental subjects were compared against 31 control subjects for choice change between pretest and posttest conditions, The experimental subjects were placed into four groups, which represented approximate quartile divisions of the I-E Scale and a comparison of the frequency of psychological reactance in the four quartiles was made.Previous studies in this area had used a mean or a median split on the I-E Scale or had used the extremes of the scale.Two null hypotheses were tested by using the Standard Normal Test and by using Chi-Square, The .05 level of significance was established as the critical probability level for the non-acceptance of hypotheses.Findings1) Reactance was exhibited by the experimental subjects. 2) Differences in reactance were not significant when compared by quartiles.ConclusionsBrehm's (1966) theory of psychological reactance was supported. Me findings do not support the conclusions of Cherulink and Citrin (1974).On the basis of the findings, conclusions were drawn and speculations were made concerning the utilization of paradoxical psychotherapy.

Many large-scale national and international testing programs use the Rasch model to govern the construction of measurement scales that can be used to monitor standards of performance and monitor performance over time. A significant issue that arises in such programs is that once a decision has been made to use the model, it is not possible to reverse the decision if the data do not fit the model. There are two levels of question that result from such a situation. One of them involves the issue of misfit to the model. That is, how robust is the model to violations of fit of the data to the model? A second question emerges from the premise that the issue of fit to the model is a relative matter. That is, ultimately, it becomes the decision of users as to whether data fit the model well enough to suit the purpose of the users. Once this decision has been made, such as in the case of large-scale testing programs like the ones refocused to above, then the question reverts to one in which the focus is on the applications of the Rasch model. More specifically, in the case of this study, the intention is to examine the consequences of variability of fit to the Rasch model on the measures of student performance obtained from two different equating procedures. Two related simulation studies have been conducted to compare the results obtained from using two different equating procedures (namely separate and concurrent equating) with the Rasch Simple Logistic model, as data-model fit gets progressively worse. The results indicate that when data-model fit ranges from good fit to average fit (MNSQ ??? 1.60), there is little or no difference between the results obtained from the different equating procedures. However, when data-model fit ranges from relatively poor fit to poor fit (MNSQ &gt 1.60), the results from using different equating procedures prove less comparable. When the results of these two simulation studies are translated to a situation in Australia, for example, where different states use different equating procedures to generate a single comparable score and then these scores are used to compare performances amongst students and to predetermined standards or benchmarks, it raises significant equity issues. In essence, it means that in the latter situation, some students are deemed to be either above or below the standards purely as a consequence of the equating procedure selected. For example, students could be deemed to be above a benchmark if separate equating was used to produce the scale; yet these same students could be deemed to fall below the benchmark if concurrent equating is used. The actual consequences of this decision will vary from situation to situation. For example, if the same equating procedure was used each year to equate the data to form a single scale, then it could be argued that it does not matter if the results vary from occasion to occasion because it is consistent for the cohort of students from year to year. However, if other states or countries, for example, use a different equating procedure and the results are compared, then there is an equity problem. The extent of the problem is dependent upon the robustness of the model to varying degrees of misfit.

Empirical evidence suggesting colour influences object recognition is mixed; leading to conclusions that colour may not always be represented in object memory. Positive evidence for the representation of colour in episodic object memory is often complicated by the possibility that encoding specificity may be responsible for such observations. The current thesis examined whether colour is represented and makes an independent contribution of shape in episodic memory for familiar and novel objects, using a modified paradigm based on the typical retrieval- practice task (e.g., Anderson, Bjork, & Bjork, 1994). Participants studied pictures of objects, presented one at a time. In a subsequent practice phase, participants either performed Old/New recognition with a subset of the studied objects and their distractors (Experiments 1-7), or they rated a subset of the studied objects for attractiveness, complexity, and usefulness (Experiments 8 and 9). The critical manipulation concerned the nature of unpracticed objects. Unpracticed objects shared either shape only (Rp- Shape), colour only (Rp-Colour), both shape and colour (Rp-Both), or neither shape nor colour (Rp-Neither), with the practiced objects. Interference effects in memory between practiced and unpracticed items are revealed m the forgetting of related unpracticed items - retrieval-induced forgetting (RIF). If both shape and colour information is explicit in the object representations in episodic memory, then there would be significant RIF for unpracticed objects sharing shape only and colour only with the practiced objects. RIF was significant for Rp-Shape and Rp-Colour objects, suggesting that shape and colour are represented and independently drive competition effects in episodic object memory. The use of RIF to probe those representations improves on previous evidence, because it bypasses alternative encoding specificity explanations. The current work provides proof of concept for a modified retrieval-practice paradigm and establishes it as a tool to probe feature- based representations that do not easily lend themselves to retrieval practice.

Prepulse inhibition (PPI) is a popular paradigm in sensorimotor gating research. In healthy individuals the weak lead stimulus (i.e., the prepulse) presentation results in a reduction in the startle probe (pulse) elicited response. The motor responses to the prepulses (prepulse reactivity, PPER) were until recently largely ignored in PPI research. There are conflicting reports about prepulse reactivity and startle response modification (SRM) associations; and personality factors relevant to SRM have not been previously examined in prepulse reactivity context. Healthy participants were drawn from university student and staff population. Three paradigms were used: unpredictable stimulus onset, predictable stimulus onset and conscious stimulus processing. The stimuli consisted of 80, 85 & 90dB prepulses and 115dB startle probe separated by 140ms inter-stimulus interval (onset to onset asynchrony). The inter-trial intervals varied between the studies. Startle responses were measured as eye blinks and recorded using surface EMG. All motor responses were quantified according to the same set of rules. Prepulse-elicited motor responses reliably appeared in all the studies and were distinct from spontaneous EMG. Some PPER characteristics exhibited stimulus intensity dependence further proving PPER validity as stimulus-driven response. Prepulse reactivity exhibited significant associations with startle response modification. PPER was a stable tendency; individuals either consistently responded to the weak lead stimuli or did not. Two types of startle response modification appeared under the conditions assumed to elicit maximal inhibition only: classical inhibition (as expected) and paradoxical prepulse facilitation. These appeared in motor responses and in conscious stimulus processing. The propensity towards the paradoxical prepulse facilitation was reduced by efficient prepulse inhibition. PPER and SRM had limited associations with personality factors, sex, or age. The predictable stimulus onset paradigm however highlighted the associations of the defensive startle response and its modification with fear and anxiety. Increased emotionality, regardless of its valence, proved detrimental to sensorimotor gating.

M.A. / New job and study opportunities were created for many people because of political changes in South Africa. This implied that many people left their homes and moved to new areas. Research indicated that 60% to 70% of first year students suffered from homesickness during their first semester at university. For some of them it was only a light experience; for others it was an intense negative experience with serious social-, academic-, and financial implications which could result in illness or even death. The role of salutogenic factors (factors that contribute to health) in the occurrence of homesickness, experienced by Afrikaans- and English speaking male and female first year students at the Rand Afrikaans Universiy, was investigated. It was found that the experience of homesickness was diminished by the salutogenic factors "sense of coherence", "self-efficacy", learned resourcefulness" and social support. These factors thus contributed to health. The factor "hardiness" did not play a salutogenic role in this study. Although the multicultural context of South Africa was not taken into account, the study should be useful in identifying students who would be prone to homesickness and illness during their first year at university.