本研究透過一顧客滿意度之議題,針對線性迴歸模型、Ordered Logit及Ordered Probit 三種模型,在應變數具順序特性時做一分析比較。研究結果發現,三種模型在實證結果並無太大的顯著差異,惟Ordered Logit/Probit模型在預測能力上明顯地優於線性迴歸模型。此外,Ordered Probit模型除了可以看出自變數影響應變數變化的方向之外,更可進一步地計算出自變數增加一單位對應變數各別尺度(非常不滿意、不滿意、普通、滿意、非常滿意)變化的邊際機率,提供行銷人員掌握消費者更豐富的滿意度資訊。因此,建議後續研究學者往後若在處理應變數為具有順序性質的類別資料時,可參考利用Ordered Probit模型,以提供研究結果更多的訊息。
This study reviews the use of Logit and Probit models in marketing and focuses on demonstrating the use of ordered probability models. A comparison between the properties of the linear regression model and ordered Logit and Probit models is made using customer satisfaction data on banking industry. This comparison between the three models shows that the results are not different significantly. However, the predicted ability of Ordered Logit and Probit is significantly better than linear regression model, and Ordered LogIt and Probit model also provide the marginal probability to let marketing managers to get more information of consumer satisfaction. Thus, the paper concludes that ordered probability models, such as the ones illustrated, should be employed in marketing and business research where the dependent variable is ordinal.