近年來，各鄉鎮縣市政府或是某些行業為了不同的目的，提供了免費公車之福利，這項福利也為民眾帶來了許多便利性；本研究將針對免費公車與一般公車的異同性為出發點，探討影響免費公車之影響因子，並且將免費公車依照功能與特性之不同分門別類；接著利用地區特性之不同，將變項聚合為總體變項，利用階層線性模式之理論，探討變項關係。 本研究收集全台灣35個鄉鎮縣市之免費公車服務品質、知覺價值、乘客滿意度與行為意向問卷資料，並且利用這四個構面，討論各構面與免費公車種類之間的關係，以及四個構面之個體與多層次中介效果與調節效果。 本研究之數據結果顯示，免費公車種類對於各個構面造成顯著影響，但對於構面之間之影響不具有調節效果；而構面之間存在著完全或部分中介效果，總體變項也能對個體變項之影響造成調節效果，但總體之間不存在著中介效果。In recent years, several types of fare-free bus services with varying service emphases have been established in many regions in Taiwan. However, the performance and effectiveness of these services have not been well understood. This master thesis research is aimed at investigating the factors affecting the perceived performance of the fare-free buses by utilizing the hierarchical linear modeling (HLM) approach. HLM is a multi-variate multi-level analysis approach that is a more advanced form of multiple linear regression models. HLM analysis allows variance in outcome variables to be analyzed at multiple hierarchical levels, whereas in multiple linear regression models all effects are modeled to occur at a single level. Thus, HLM is appropriate for use with nested data through the concept of so-called moderation and mediation effects. In this research work, fare-free buses offered in different regions were classified into different types according to their functional and service characteristics. In the HLM hierarchical structure, individual variables were aggregated by regions and represented in the higher level, and the relationships among different variables at same level as well the relationship between different levels were analyzed using HLM. This research utilized survey data collected through 1401 in-person interviews from 35 cities in Taiwan from a prior study. The questionnaire included four dimensions (variables) - service quality, perceived value, satisfaction and behavior intention. The main research questions and hypotheses studied in this research focused on revealing the relationship between these four variables and different types of fare-free buses, and the mediation and moderation among these four variables. The analysis results showed that the type of fare-free buses significantly influences these variables; however, the fare-free bus type does not significantly influence how these four variables affect every other. Moreover, it was found that there exists a partial or complete mediation effect of individual and multilevel effects among the four variables. The aggregate variables can moderate the relationship among four variables.
