DSpace Collection:http://hdl.handle.net/10722/386552024-03-28T10:21:37Z2024-03-28T10:21:37ZBayesian censoring approach to rounded zeros in compositional dataLeung, Tak-ching梁德貞http://hdl.handle.net/10722/2443352024-01-31T08:20:10Z2017-01-01T00:00:00ZTitle: Bayesian censoring approach to rounded zeros in compositional data
Authors: Leung, Tak-ching; 梁德貞
Abstract: The logratio transformation is commonly used for analyzing compositional data. However, logratio analysis is not possible when the data has any zeros. However, zeros are common in data on proportions, concentrations etc. when the data is small enough to be below the detection limit, i.e. left-censored data.
Various methods to handle zeros such as combining data and replacing the zeros with predicted non-zero values have been proposed, including additive and multiplicative replacements. The replacement value can be fixed or variable, and can be obtained from various iterative means like modified EM algorithm.
In this thesis, Bayesian analysis assuming left, right and interval censoring on the original scale of compositional data is proposed and applied to the 3 different data sets and simulated data. This approach has the advantages of reflecting the data recording process, which necessarily involves censoring, performing well on the simulated data and yielding easy interpretation of results.
The new approach is demonstrated by means of models estimated using the standard Markov Chain Monte Carlo(MCMC) package WinBUGS. Pioneering models were also estimated using the new Hamiltonian Monte Carlo(HMC) package, RStan. Unlike the models in WinBUGS and all previous approaches in the literature, the Rstan models are based on censoring over the correct region in the simplex, rather than an approximate censoring region based on the log-ratio space. Models from both tools showed reasonable and consistent results. The advantages and limitations of them are compared and discussed.
Apart from zeros, which are left-censored, both methods can handle right-censored and interval censored data as well.
The Rstan models are particularly promising. They are built on a simple architecture which uses numerical approximation of the integral of probabilities over the censored region. It is comparatively fast and flexible, allowing for the possibilities of modification with different ways to handle the approximation.2017-01-01T00:00:00ZFirst- and third-person perceptions in health communication by the Hong Kong mediaChen, Shishi陳什什http://hdl.handle.net/10722/2406522020-08-08T10:52:12Z2017-01-01T00:00:00ZTitle: First- and third-person perceptions in health communication by the Hong Kong media
Authors: Chen, Shishi; 陳什什
Abstract: Since health-related messages are in every kind of media that one might use in a typical day, it has become clearer that the media indeed shapes individuals’ health-related beliefs and behaviors. Negative health communication is a major area of interest within the study of the third-person effect (TPE). In recent years, interest has been increasing in the first-person effect (FPE) in positive health communication. Hitherto, there has been little discussion about the segmentation of potential audiences by their attitudes toward positive health information. Moreover, there is no relevant research to date in Hong Kong.
This thesis discusses the perceptual component of both FPE and TPE, namely first-person perception (FPP) and third-person perception (TPP) respectively. It explores which factors can affect the perceived media influence from both a psychometric and empirical perspective. It aims to investigate the self-other perceptual impact disparity as shown through the FPP and TPP toward health information released by the government and corresponding commercial adverts among local college students from different disciplines. Structural equation modeling is used to conceptually analyze different health topics.
Message attributes is found to be the only factor that always influences self-other perceptual impact disparities toward all adverts and that applies to all the respondents despite their different health-related knowledge. Message attributes is always positively correlated with perceptual impact on self. However, the relationships of message attributes and perceptual impact on other students are varied which depend on different health topic. Respondents show FPP from desirable adverts and TPP toward undesirable adverts.
Although other factors are only found to affect the perceptual impact on self and others toward desirable adverts, they do not cause a significant difference in FPP or TPP. Females and part-time Nursing students with health-related work experience are more likely to perceive positive influence on both self and other students from desirable adverts. Frequently accessing non-official health-oriented websites also leads to more positive perceptual impact on self and others.
No perceptual impact difference is seen between Medicine and Dentistry students who are trained with professional health education and other students from non-clinical disciplines. Comparison within Medicine and Dentistry students indicates that senior students who acquire more professional health-related knowledge show either stronger TPP or weaker FPP. However, in terms of topic teeth care, Dentistry freshmen always perceive stronger FPP than other non-dental students, which may be because their perceived dental knowledge is undervalued.
No clear pattern between health condition and perceptual impact towards desirable adverts can be found. No factor can be found that affects the perceptual impact toward undesirable health-related adverts either.
This is the first study to undertake a comparison among different disciplines of college students, and the findings make an important contribution to the study of positive health communication in a non-Western population. Due to practical constraints, the study topics of this research all evoke mild responses, and the perceived media influence may therefore not be as strong as that of other more intense issues.2017-01-01T00:00:00ZThinking in a tank : a comparative study on the role and impact of policy institutes in selected Asian politiesCooper, Daisyhttp://hdl.handle.net/10722/2111162020-10-08T11:58:19Z2014-01-01T00:00:00ZTitle: Thinking in a tank : a comparative study on the role and impact of policy institutes in selected Asian polities
Authors: Cooper, Daisy
Abstract: Think tanks are public policy research institutes that seek to play a role in making or influencing policy. While each think tank serves a specific purpose, they all share a common vision to improve their respective spectrum, as well as being a source of new ideas and research. However, think tanks in Asia are misunderstood, rarely acknowledged, and under-researched.
This study is a cross-polity and audience comparison, primarily using semi-structured interviews, grounded theory qualitative analysis and ethnography, to gain insight into the perceptive of think tanks, academics, media, policy makers, and chambers of commerce in Hong Kong, Malaysia and Singapore, to answer what role and impact local think tanks have on the policy making process, and to explore why.
Findings revealed six themes and 21 sub-themes. Three sub-themes emerged as having been discussed across all three polities by all five sources. The sub-themes in the order of highest reference count frequency first were: “[Think tanks have] Limited effectiveness in the policy making process” followed by “Research and dissemination [as a role]”, and “Limited media presence except for a few standout think tanks”.
Overall there were distinct differences between polities, which may, in most part, be due to varying political development in polities. There were also distinct differences between sources - possibly as the role one plays shapes their perception of issues. The source differences were generally consistent across polities, suggesting that source differences are independent of polity differences.2014-01-01T00:00:00ZChina-India interactions : implication on bordersAstarita, Claudiahttp://hdl.handle.net/10722/2103082020-08-08T10:52:12Z2010-01-01T00:00:00ZTitle: China-India interactions : implication on borders
Authors: Astarita, Claudia2010-01-01T00:00:00Z