Page 240 - Contributed Paper Session (CPS) - Volume 1

CPS1290 Sahda R. et al.
households, but significant for the white race. Conversely, Drew (2014) states
that work does not affect home ownership.
The larger household members who live together, the more space needed.
Houses must be able to meet the needs of all household members. Aizawa
and Helble (2016), states that the number of household members is the
strongest variable affecting Japanese people to become homeowners than
income variable. Likewise with the research of Guris, Caglayan, and Un (2011)
which found that the type of extended family and couples without children
significantly influenced the decision to own a house. But in the Skak and
Lauridsen study (2007) the number of household members did not
significantly increase the chances of a family owning a home. Fisher and Jeffe
(2003) found that in macro terms, the number of household members did not
affect the homeownership rate in a country.
Constant, Roberts, and Zimmermann (2007), Tan (2008), Drew (2014) also
show that the existence of children influences the reason to own a home.
According to Gendelman (2005), the significant influence is not the presence
of children, but the number of children under the age of 18 in the household.
On the contrary, Lauridsen and Skak (2007) found that children's existence did
not significantly influence the decision to housing ownership in Denmark.
This study, however, tries to elaborate those factors as variables
determining housing ownership in DKI Jakarta. By doing so, it can be a
reference for the local government to evaluate variable of inclusive
housingownership program.

2. Methodology
The data used in this study is sourced from the 2017 National Socio-
Economic Survey (SUSENAS) by the BPS-Statistics Indonesia by using 5062
household samples. Probit model is used to analyse the determinants of
housing ownership. The dependent variable in this study is home ownership
status which is a dichotomous variable which are homeowner and not
homeowner. Meanwhile, the independent variable to explain the dependent
variable consists of eight variables, i.e. household expenditure, gender, age,
education level, marital status, and employment status of the head of the
household, number of household members, and the presence of children.
The dependent variable is a categorical dichotomous (binary) variable, so
there are 3 models that can be used, namely the logit model, the probit model,
and the gompit model (complementary log-log). Logit models and probit
models can be used if the dependent variable data tends to be symmetrical,
or the amount of data between categories is almost the same. Conversely, the
gompit model is used for not symmetrical data (Agresti, 1990).
In this study, the model used is a probit model, because the comparison
of data between categories on the dependent variable is quite

229 | I S I W S C 2 0 1 9