Page 216 - Contributed Paper Session (CPS) - Volume 2

CPS1825 Suryo A.R.
correlation between general factor formed and each indicator. The bigger lij
means the bigger correlation between them.
There are some steps in factor analysis. First, identify the purpose of using
factor analysis and fulfil its requirements. Secondly, checking the correlation
matrix in two ways. Bartlett test of sphericity and measuring Keiser-Meyers-
Oklin (KMO) or Measure of Sampling Adequate (MSA) to assess the data
appropriateness. The third step is factor extraction with methods principal
component analysis. The fourth step is factor rotation, and the last is getting
factors score to construct the index.
Data used in this research are from Socio-Economic National Survey
Indonesia (SUSENAS) conducted by BPS-Statistics Indonesia. All the indicators
used are based on the survey in 2018 so this index will capture the
development of youth in 2018.

3. Result
Construction of Youth Development Index of South Kalimantan 2018

Indicator Selection and factor construction
The selection process of indicators uses anti-image matrix to decide
whether an indicator deserves to be analysed further or not. The cut point of
MSA score is 0,5. If the MSA score of an indicator more than 0,5, it means that
the indicator deserves to be analysed further in factor analysis.
In the first step, indicator vocational experience and literacy in education
dimension and indicator access for the youth migrant in gender and
discrimination must be reduced because they have the lowest MSA which are
less than 0,5. It means that the other three dimensions have indicators which
are good to be analysed without any reduction. After the indicators having
MSA less than 0,5 are reduced, the education dimension and gender and
discrimination dimension are good to be analysed using factor analysis.
Factor analysis will produce dominant factors in every dimension of the
index. The number of dominant factors which characterize the dimension
could be decided based on the Kaiser criteria. Kaiser criteria is when factor
whose eigen value is more than one would be the dominant factor (OECD,
2008). The factor score would be the dimension score in this research, because
every dimension has one factor left after rotating the component score. The
dominant indicators in every dimension are mean years of schooling,
morbidity, white collar, participating on the forum, and access for the
disability.
The most highlighted new indicator, access for the disability can be the
most dominant indicator for gender and participation dimension. It means
that access for the disability has an impact in the dimension which can reflect
one of the aspects used for government evaluate and plan the development

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