Class of 2016 Abstract / Objectives: To determine the number of patients treated with eribulin who required an alternate dosing schedule other than “day 1/day 8” due to side effects. Methods: Chart reviews were conducted on all patients who met inclusion criteria. Data collected included patient demographics, history of surgery/radiation, number of past chemotherapy treatments, and lab values prior to each eribulin cycle. Results: A total of 37 patients met inclusion criteria for this study. Ten patients were initially started on the “day 1/day 8” schedule and 3 of those patients required a change to the extended “day 1/day 15” schedule. The remaining 27 patients were started on the extended schedule. Conclusions: The number of patients requiring a dosing schedule change due to side effects was not statistically significant. This finding was due to the fact that the majority of patients were started on an alternate dosing schedule in the beginning of treatment. More extensive studies would be required to determine if a majority of patients would require this alternate dosing schedule, and if this should be initiated in all patients starting on eribulin.

Ce travail doctoral a eu pour but de comprendre le rôle dynamique des moments de différentes intensités et valences perçues aux trois périodes d'une expérience de service (début, milieu, fin) sur la satisfaction globale dans le temps. Une revue de la littérature a été effectuée, s'articulant autour de quatre axes de recherche : l'expérience de service, la satisfaction du client, l'évaluation rétrospective d'une séquence temporelle et la mémoire, combinée alternativement à l'étude qualitative. Notre étude empirique comporte deux phases. Tout d'abord, une étude préalable a été réalisée par la méthode expérimentale en laboratoire comprenant 3 sous pré-tests. L'étude a eu pour but d'identifier 12 événements de services qui constituent ensuite une expérience (un festival) composée de trois périodes dont chacune comprend 4 différents événements (4 moments). L'étude principale a traité de ce festival en contexte réel (trois expérimentations, au Vietnam) en manipulant trois dimensions (saillance, valence et position temporelle de l'événement) avec une mesure longitudinale de la satisfaction globale (immédiatement après l'expérience, trois semaines après et six mois plus tard). Les tests des hypothèses sont réalisés à partir d'ANOVA-mesure répétée, d'équations structurelles, et d'analyse de dominance / This doctoral work attempts to understand the dynamic role of different moments within a service experience having different intensities, valences and existing across different service periods (start, middle, and end) on global satisfaction levels within a longitudinal perspective. The literature review which focuses on four streams: service experience, client satisfaction, retrospective evaluation of a temporal sequence, and memorization, was conducted alongside a qualitative study. It was followed by our empirical study which has two phases. The first phase is the preliminary studies in the laboratory wherein 3 pre-tests were conducted in order to identify 12 different moments of a service experience suitable for enactment in the final study as a real life festival. Moreover, the structure of the festival was also selected from the results of these pre-tests. In the second phase, a real life festival (3 different experiments, in Vietnam) was organized and the three dimensions (intensity, valence and temporal position) of the moments were manipulated. Additionally, a longitudinal measurement of global satisfaction (immediately after the festival, 3 weeks after the festival, 6 months after the festival) was taken. The data was subsequently treated with repeated measures ANOVA, structural equation modeling and dominance analysis. Results show differential effect of the moments on global satisfaction over time depending on the different levels of intensity, valence and temporal position

影響選民投票抉擇的變數可以分成長期因子與短期因子，長期因子有政黨認同與對於現任政府施政評價等；短期因子則包括對於候選人的評價以及競選期間所發生的事件等。先前的研究指出，當一個選民的長期因子與短期因子出現衝突，則該選民將可能不會投給較偏好的候選人。和這些研究不同的是，本論文從衝突選民的理論假設出發來分析投票參與行為。本論文認為，長期因子與短期因子若出現衝突，會使得選民產生衝突壓力，無法作出判斷，進而降低選民投票的意願。本論文聚焦「政黨認同」以及「對現任政府的回溯性評價」兩個長期因子，以及「對候選人的偏好」做為短期因子總指標，探討長期因子與短期因子發生衝突時對於選民投票參與的影響。本論文根據政黨認同與候選人偏好一致與否，以及回溯性評價與候選人偏好一致與否，將選民做若干分類進行分析。本論文利用「台灣選舉與民主化調查」 2004年、2008年、2012年與2016年總統選舉調查資料進行實證分析，發現當選民的政黨認同與候選人偏好並非一致時，選民的投票意願降低。其次，回溯性投票與候選人偏好的衝突也有相似的效果，對於執政黨的表現滿意與否，若非與候選人偏好一致，選民的投票參與意願也會下降。整體而言，本論文將衝突選民理論拓展到對於投票參與行為的解釋，除了對於投票行為研究具有若干貢獻外，也對於政黨的選舉策略提供一些啟發。 / Many previous studies have tested the conflicted voter hypothesis that voters will be less likely to vote for their preferred candidate when the voters' issue positions are in conflict with their partisanship. Building on these previous works, this thesis tests the conflicted voter hypothesis on electoral participation. I contend that voters are less likely to vote when they are under pressure generated by conflicts between factors. Using Taiwan's Election and Democratization Study (TEDS) surveys for the 2004, 2008, 2012, and 2016 presidential elections in Taiwan, the empirical analysis demonstrates that voters are less likely to vote when their partisanship is not consistent with their preferred candidate's partisanship. Moreover, the analysis also shows that voters tend to not vote when they identify with the opposition party but are satisfied with the incumbent government's performance, and when they identify with the incumbent party but are dissatisfied with the incumbent government's performance. Overall, this paper not only facilitates a better understanding of conflicted voters' behavior, but also provides implications for the campaign strategies of political parties.