Abstract:
The 2020 school participation figure states that 20.56% of children in the disability
category have the status of not/never been to school (BPS, 2020). This shows that there
are still many children with disabilities who have not received adequate education.
Therefore, attention to the availability of facilities and access to education for children
with disabilities needs to be increased so that there is no inequality of school
participation between children with disabilities and non-disabled children. On
extraordinary school data statistical methods can be applied for various purposes. The
method that can be used to group mixed-type data is ensemble. In this study, the
ensemble ROCK (Robust Clustering using links) method was used at 47 extraordinary
schools in Riau Province. Using the value 𝜃 of 0.22 in the ROCK ensemble method, we
get 3 optimal clusters with a ratio of 0.08177794. It was found that cluster 3 is a cluster
that does not have adequate facilities such as a lab, library and internet network than
other clusters. It can be said that cluster 3 needs more attention than other clusters.
Keywords: Disability, extraordinary school, mixed type data, ensemble ROCK.
