Akademik Veri Yönetim Sistemi
Tezin Türü: Doktora
Tezin Yürütüldüğü Kurum: Orta Doğu Teknik Üniversitesi, Mühendislik Fakültesi, Endüstri Mühendisliği Bölümü, Türkiye
Tezin Onay Tarihi: 2013
Öğrenci: AYŞEGÜL ALTABAN
Danışman: GÜLSER KÖKSAL
Özet:A major interest of durable goods' producers is factors affecting satisfaction and loyalty of their customers. However, this topic is covered in the literature to a limited extent. Existing studies focus on use of structural equation modeling for studying loyalty for cars, white goods. These are developed as general perception models to serve as customer satisfaction index models for Turkish and also for global consumers. In this study, we also develop a perception model for another durable good, residential air conditioner. However, our model is much more comprehensive as to the number and scope of modeling variables. Items on technical features are used together with perception questionnaire items. Thus, consumers’ technical experiences are combined with their consumption experiences and with their relations with vendors. In the existing literature, factors affecting consumption of long-lasting goods are studied using factor analytic approaches. Factor analysis is a small structural equation modeling application and does not include latent paths (structural regression equations). Thus it is just a confirmatory tool. Our model is a full structural equation model with factor analysis and also latent paths. On the other hand, inherent influential variables are not incorporated in existing models. We model customer perceptions for air-conditioners and we use more factors (latent variables) than those of the existing studies (on both goods and services). We also enrich our model with three covariates; length of relationship, education and income. In our model, “length of relationship” is studied as the major covariate in explaining long- term consumer attitudes. This variable is studied as the major explanatory variable in our structural models. Interactions of length of relationship with attitude factors are also included in the models. Regression, moderation and latent variable interaction techniques are used to model interactions.