交岔路口左轉車輛讓對向直行車輛先行意向之研究

Abstract

本研究嘗試從「駕駛人心理」的角度出發探討左轉車讓對向車先行之意向，構建左轉讓車之駕駛意向模式，以找出影響左轉讓車意願之因素；此外，本研究也設計駕駛情境試題，分析駕駛者於各駕駛情境下讓車意願之差異，期能找出「發生搶左轉機率較高的駕駛情境」。至於本研究使用的研究方法乃使用問卷調查法、徑路分析法、Rasch分析法、集群分析法等，其中問卷設計乃依據本研究所假設之駕駛意向模式進行設計，模式中包含「人格特質」、「駕駛技術自信」、「風險感認」、「讓車意願」等四大構面，每一構面皆設計若干題目進行衡量，然而問卷中所使用的「李克度尺度(likert-scale)」存在「不等距」的爭議，因此導入試題反應理論中的Rasch分析以克服該問題。本研究蒐集全國有騎乘機車或駕駛汽車經驗者之樣本，經由預試修改問卷，正式調查共回收479份樣本，有效問卷386份，並利用Rasch模式對原始資料進行尺度轉換，利用各樣本資料於各構面的「Rasch Score」進行後續模式構建分析與統計分析。研究結果顯示，搶先左轉的駕駛人並不覺得「左轉車與對向直行車輛衝突」狀況會威脅到他們的自身安全。另外，對機車駕駛者而言，較無遵守法規觀念(無規範感較高)之駕駛者，有較高的搶先左轉的可能性。除此之外，對在各族群的左轉車駕駛者而言，當對向車減速接近但尚未抵達停止線時，普遍有較高的機率搶先左轉。綜合以上，本研究認為，由於影響左轉讓車意願最大因素為「風險感認」，再加上搶先左轉駕駛者無視交通法規，因此無論對這些不守法的駕駛者執行許多取締手段，其效果有限，因此提昇駕駛人面臨左轉車與對向直行車衝突狀況時的「恐懼程度」應為改善搶先左轉現象之重點策略，要使駕駛人覺得「搶先左轉」這個行為可能導致非常嚴重的後果，應可減低搶先左轉的現象；此外，駕駛訓練時就應實施情境訓練，教育駕駛者在較危險的駕駛狀況時，應不要冒險搶先左轉。本研究可提供相關單位於「雙向雙車道非號誌化交岔路口」之搶先左轉現象改善策略擬定之參考。

This study explores drivers’ psychological intentions while yielding to opposite straight going cars at intersections. A model was constructed to find factors related to yielding to opposite straight going cars; In addition, the study designed some driving scenario questions to analyze drivers’ willingness to yield at all traffic condition, hoping to find "the scenario with the highest rate of occurrence of looting to opposite straight going cars". This study used some methods such as questionnaire surveys, path analyses, cluster analysis and the Rasch model. Based on the theory of driver intention, the questionnaire was designed using personality traits, driver skill confidence levels, risk recognition, and willingness to yield to straight going cars. Every construct was measured by several items. Often, in other similar studies, the Likert-scale is used as the interval scale. However, because there is no evidence to prove that the Likert-scale is the correct interval scale for this type of research, and the Likert-scale is really just an ordinal scale, it is not used for statistical analyses. To overcome this dispute, this study tries to apply item response theory to convert ordinal data into interval data for further statistical inference and discussion. This study collected driving samples from both motocycles and automobiles. In the formal testing, 479 samples were collected, with 386 valid samples. The results demonstrate that risk recognition is the main factor effecting driver behavior as related to failing to yielding to opposite straight cars. It means that drivers who often fail to appropriately yield to opposite straight going cars do not feel dangerous when they meet the traffic conflict. Motorcyclists who has the high normlessness level may loot to opposite straight going cars. In addition, regardless of what the left turning car is, the left- turning drivers had the high probability to loot the opposite straight going car which had not yet arrived at the stop line. According to the results, this study suggests that increasing driver awareness of the risk recognition of failing to yield to opposite straight going car at intersections is an important issue for the government to mitigate the phenomenon of failing to yield to opposite straight going cars at intersections.