The mean residual life (MRL) function is an important function in survival analysis, actuarial science, economics and other social sciences and reliability for characterizing lifetime. Different methods have been proposed for doing inference on the MRL but their coverage probabilities for small sample sizes are not good enough. In this thesis we apply the empirical likelihood method and carry out a simulation study of the MRL function using different statistical distributions. The simulation study does a comparison of the empirical likelihood method and the normal approximation method. The comparisons are based on the average lengths of confidence intervals and coverage probabilities. We also did comparisons based on median lengths of confidence intervals for the MRL. We found that the empirical likelihood method gives better coverage probability and shorter confidence intervals than the normal approximation method for almost all the distributions that we considered. Applying the two methods to real data we also found that the empirical likelihood method gives thinner pointwise confidence bands.

In Tanzania, an ongoing private media boom operates within the frameworks of media laws dating back to the 1970s, restricting media freedom and enabling the government to keep some control of what is written on certain topics. This is widely acknowledged and makes Tanzania an interesting subject for studies of news media confidence from an audience perspective. Confidence is viewed as an attitude consisting of three components: the cognitive, the emotive and the behavioral component. The cognitive component consists of expectations and sought gratifications, the emotive component is the feeling of confidence or the disappointment of a failed expectation and the behavioral component includes what media an individual intends to use, which should not be mistaken for what media an individual is actually using. A quantitative questionnaire in Swahili was distributed to 84 randomly selected inhabitants of Dar es Salaam, almost equally distributed between three areas: the lower class area Mwanayamala, the middle class area Kimara and the upper class area Mikocheni. After finishing the quantitative study, 9 qualitative interviews were conducted with a representative selection of those answering the questionnaire. The interviews and questionnaires were conducted during November and December 2013 and all participants were guaranteed anonymity. Finding truthful information, getting education and entertainment was the most frequent sought gratifications from media usage. Media type, language and ownership were the medium characteristics most evidently affecting confidence. Religion, social status and interests were the individual characteristics most evidently affecting confidence. News media confidence turns out to be more about mutual understanding between the individual receiver and the sending medium, than perceived accuracy and independent monitoring of power.

Thesis (M.A.) - Carleton University, 2005. / Includes bibliographical references (p. 58-62). Also available in electronic format on the Internet.

The confidence interval estimate of percentile and its applications were studied. The three methods of estimating a confidence interval were introduced. Some properties of order statistics were reviewed. The Monte Carlo Method -- used to estimate the confidence interval was the most important one among the three methods. The generation of ordered random variables and the estimation of parameters were discussed clearly. The comparison of the three methods showed that the Monte Carlo method would always work, but the K-S and the simplified methods would not.

Indiana University-Purdue University Indianapolis (IUPUI) / The number of sports officials around the United States and world has been slowly declining. The average age of sports officials keeps rising year over year, meaning younger officials are leaving their jobs at a high rate. The goal of this study was to compare training methods, traditional and immersive video-based, specifically RefReps training, to see how this training method affects officials’ confidence level at performing their duty as college intramural basketball officials. The study included college level intramural basketball officials from different universities in the United States (N = 34). A pre-test survey asking about confidence level was administered to each participating official prior to training for the upcoming intramural season. A control group (N = 13) used the university’s traditional training as their only form of training, while the test group (N = 21) trained using RefReps immersive video-based training. A post-test survey was given with the same confidence level questions for the participants to answer. The results showed that the RefReps group showed a greater increase in confidence in their ability to possess the most extensive knowledge of the rules compared to anyone else at the game and an increase in confidence in their ability to determine if a foul or violation had been committed.

A self-handicapping conceptualization of the function of the self-applied codependent label is presented. It was proposed that the self-appellation of the codependent label would function as a self-handicap when used by women who were not children of an alcoholic (COAs). It was also proposed that such a use of this label constituted a tendency to self-handicap and that these women would be more likely to employ self-handicapping strategies in situations involving interpersonal evaluations than: (1) controls who are not COAs or codependent; (2) COAs who describe themselves as codependent; or (3) COAs who do not endorse codependency. Additionally, in situations which involved interpersonal evaluations by males portrayed as either exploitive or nurturant, it was predicted that women who were COAs (with or without the codependent label) would likely to rate the exploitive male positively. The results did not support the self-handicapping hypothesis for women who labeled themselves codependent. The evidence suggests that COAs may self-handicap more than non-COAs. In addition, those women who endorsed the codependent label regardless of COA status did not distinguish between exploitive and nurturant males in liking, and non-codependent COAs liked the exploitive male least.

Item response theory (IRT) offers several advantages compared to classical test theory (CTT) in providing additional information on psychometric qualities of the scale. My goal was to demonstrate the superiority of IRT as compared to CTT through two analyses of the Top Leadership Direction scale (TLDS), which was created to measure the effectiveness of top leadership through the followers’ perceptions in the context of providing guidance of the organization. Furthermore, the participants (n = 8046) were the employees from various positions at 18 of the 23 California State University campuses. In the graded response model (GRM) analysis, the result showed that IRT provided more information about each item and allowed a useful visual inspection of the items. With the second analysis, I aimed to provide evidence of measurement equivalence across functional groups of employees using differential item functioning (DIF) analysis in IRT. Due to the lack of model fit, the DIF analysis was incomplete. A supplementary multigroup CFA was conducted to investigate the structural difference across the groups for the items of the TLDS. The result of multigroup CFA suggested that item 2 and item 4 did not show measurement equivalence across the groups at the construct level. An alternative model in IRT was discussed due to some limitations of GRM in the present study. Practical and theoretical implications for the use of IRT were also presented and contrasted with CTT.

We propose a novel feature reduction approach to group words hierarchically into clusters which can then be used as new features for document classification. Initially, each word constitutes a cluster. We calculate the mutual confidence between any two different words. The pair of clusters containing the two words with the highest mutual confidence are combined into a new cluster. This process of merging is iterated until all the mutual confidences between the un-processed pair of words are smaller than a predefined threshold or only one cluster exists. In this way, a hierarchy of word clusters is obtained. The user can decide the clusters, from a certain level, to be used as new features for document classification. Experimental results have shown that our method can perform better than other methods.

Attitudes and confidence among women can predict the duration of exclusive breastfeeding. The longer duration of breastfeeding, the more advantages there are for both mother and child. Aim: The aim of this study was to explore how the attitudes and the confidence were among Thai mothers and if there was a difference between nuclear families and extended families in these issues. Method: A descriptive comparative cross-sectional study with a quantitative method was performed by using IIFAS (Iowa Infant Feeding Attitude Scale) and BSES (Breastfeeding Self-Efficacy Scale) as measurement tools. Orem´s self-care theory was used as a theoretical framework with 79 voluntary participating women at a well-baby clinic in Bang Pa-In, Ayudhaya, Thailand. Result: The outcome of the study demonstrated that the majority of the women had a neutral attitude towards breastfeeding, 13% had a positive attitude towards formula feeding and 7% had a positive attitude towards breastfeeding. No significance was found between mothers living with nuclear and extended family regarding the attitude. The majority rated high confidence in breastfeeding. Mothers who were living in nuclear families rated lower than mothers living in extended families in “refrain from bottle-feeding for the first 4 weeks” regarding confidence in breastfeeding. Conclusion: The attitude to and the confidence in breastfeeding were relatively good among Thai mothers in Bang Pa-In, Ayudhaya, however improvements are needed. The findings of this study could be useful for Swedish and Thai nurses in their information about breastfeeding to women. Further studies are needed to follow the attitudes and confidence along with the economic development for the improvement of the exclusive breastfeeding rates.

Our understanding of parliamentary politics suggests that no-confidence motions have a critical place in government continuation, reorganization and termination. More specifically, we know that opposition parties use no-confidence motions as a way of removing the government and potentially inducing early elections. Up until now, we know little about either the causes or the consequences of no-confidence motions. In this dissertation, I first develop a formal model of the conditions under which an opposition party will threaten to propose (and eventually propose) a no-confidence motion in the government. The model provides a number of intuitive observations about the behavior of opposition parties and the reactions of governments to challenges. I develop a competence-based theory where opposition parties signal their perception of the government's competence with no-confidence motions. In the game, opposition parties act both in terms of short-term gains as well as long-term electoral gains. This model provides intuitive answers that help us understand the circumstances under which the opposition will challenge the government. The model also provides empirical expectations regarding the probability that the motion is successful, in addition to its long-term electoral consequences. Next, I test the theoretical propositions regarding the occurrence of noconfidence motions on a cross-sectional time-series data set of all no-confidence motions in a sample of parliamentary democracies in the post-World War II era. Even though successful no-confidence motions are relatively rare, they can have profound consequences on policy outcomes. The next section illustrates these consequences, as I find that having a no-confidence motion proposed against them makes governments more likely to be targeted by other states in international conflicts. In the conclusion I summarize the key findings, present the broad implications for the study of parliamentary decision making, and discuss avenues for future research.