[Masalah utama pada penelitian ini adalah terbatasnya lahan di Indonesia khususnya wilayah perkotaan dan perumahan yang didominasi oleh rumah tapak yang memakan banyak lahan. Menurut Survei Sosial Ekonomi Nasional (SUSENAS) 1989 ? 2007 rumah tapak cenderung mendominasi dan bahkan mengalami kenaikan dari 75.36% di 1989 menjadi 86.91% di 2007 dibandingkan jenis rumah lainnya. Tentunya hal ini akan mengurangi penggunaan lahan lain seperti untuk keperluan pertanian ataupun pembangunan infrastruktur seperti transportasi publik. Penelitian ini bertujuan untuk melihat pengaruh faktor lokasi serta faktor sosial ekonomi rumah tangga terhadap preferensi status kepemilikan dan tipe bangunan rumah. Penelitian ini menggunakan data SUSENAS 2007 khusus modul kor serta perumahan dan lingkungan. Metode estimasi yang diapakai adalah regresi multinomial logit dan regresi tobit untuk mengatasi kelemahan multinomial logit. Hasil regresi multinomial logit menunjukan bahwa umur kepala rumah tangga, jumlah anggota keluarga, total pengeluaran rumah tangga bulanan, tingkat pendidikan kepala rumah tangga, keapal rumah tangga bekerja di sektor jasa dan industri, adanya anak berumur 9 tahun atau kurang, jarak ke pasar tradisional dan jarak ke kantor berpengaruh signifikan dalam beberapa kemungkinan memiliki atau menyewa rumah tapak dan rumah vertikal. Akan tetapi, variabel pendidikan tertinggi kepala rumah tangga dan jarak ke toko hanya signifikan pada beberapa kategori di regresi multinomial logit.

---

Limited land in Indonesia, especially urban areas and housing which is dominated by landed house that takes up a lot of land. According to the National Socioeconomic Survey (SUSENAS) 1989-2007, landed houses tend to dominate and even increased from 75.36% in 1989 to 86.91% in 2007 compared to other types of homes. Surely this will reduce land use for agricultural purposes, construction or infrastructure such as public transport. This study aims to look at the influence of the location as well as socio-economic factors on the preferences of domestic ownership status and the type of house building. This study uses data CORE SUSENAS 2007 as well as MODULE SUSENAS 2007 which is special for housing and the environment. Estimation method that is used is multinomial logistic regression and Tobin logistic regression to overcome the weaknesses of multinomial logistic. Multinomial logistic regression results showed that the age of family head, the number of family members, total monthly household expenses, the main work status and field of household head, the distance to the traditional markets and the distance to the office have significant effect in some possibility to own or rent a landed house or vertical house. In spite of those variables, there are some significant variables such as highest education of household head, the distance to supermarkets or stores. Nevertheless, those variables are just significant in several categories in multinomial logistic regression.;Limited land in Indonesia, especially urban areas and housing which is dominated by landed house that takes up a lot of land. According to the National Socioeconomic Survey (SUSENAS) 1989-2007, landed houses tend to dominate and even increased from 75.36% in 1989 to 86.91% in 2007 compared to other types of homes. Surely this will reduce land use for agricultural purposes, construction or infrastructure such as public transport. This study aims to look at the influence of the location as well as socio-economic factors on the preferences of domestic ownership status and the type of house building. This study uses data CORE SUSENAS 2007 as well as MODULE SUSENAS 2007 which is special for housing and the environment. Estimation method that is used is multinomial logistic regression and Tobin logistic regression to overcome the weaknesses of multinomial logistic. Multinomial logistic regression results showed that the age of family head, the number of family members, total monthly household expenses, the main work status and field of household head, the distance to the traditional markets and the distance to the office have significant effect in some possibility to own or rent a landed house or vertical house. In spite of those variables, there are some significant variables such as highest education of household head, the distance to supermarkets or stores. Nevertheless, those variables are just significant in several categories in multinomial logistic regression., Limited land in Indonesia, especially urban areas and housing which is dominated by landed house that takes up a lot of land. According to the National Socioeconomic Survey (SUSENAS) 1989-2007, landed houses tend to dominate and even increased from 75.36% in 1989 to 86.91% in 2007 compared to other types of homes. Surely this will reduce land use for agricultural purposes, construction or infrastructure such as public transport. This study aims to look at the influence of the location as well as socio-economic factors on the preferences of domestic ownership status and the type of house building. This study uses data CORE SUSENAS 2007 as well as MODULE SUSENAS 2007 which is special for housing and the environment. Estimation method that is used is multinomial logistic regression and Tobin logistic regression to overcome the weaknesses of multinomial logistic. Multinomial logistic regression results showed that the age of family head, the number of family members, total monthly household expenses, the main work status and field of household head, the distance to the traditional markets and the distance to the office have significant effect in some possibility to own or rent a landed house or vertical house. In spite of those variables, there are some significant variables such as highest education of household head, the distance to supermarkets or stores. Nevertheless, those variables are just significant in several categories in multinomial logistic regression.]