Page 408 - Special Topic Session (STS) - Volume 4

STS2320 Mohamed A I.
to factor analysis. Equal weights were used in the absence of any clear
evidence of a diversity of significance among variables, as well as in the
absence of sound and complete information concerning the existence of
causal relationships, or where a lack of consensus exists on a classical method
for estimating weights.
The budget allocation method was also used for weighting where a group
of specialists and experienced experts were invited to attend a workshop for
each of the knowledge sectors. Each expert was given a budget consisting of
100 points to award to the variables. If the variable was believed to have
greater relative importance, it was allocated a greater number of points.
Subsequently, the weights were calculated according to the average of the
total points allocated to each variable.
The weights were also assessed using factor analysis, which is based on
aggregating the linked sub-indicators to form a single factor containing as
much information as possible that is shared between these linked indicators.
The weights produced by using both the budget allocation and factor analysis
methods were consistent with each other and with the initial weights
estimates, based on the intellectual and conceptual framework, OECD (2008).

Index calculation
The 2018 GKI was calculated for 134 countries in this second edition, using
the most recent and best available data to calculate the variables for each
country, with 2007 as a cut-off year and 2006 being exceptionally used for
specific countries that required additional data to qualify for inclusion in the
Index. The values of the composite sub-index were calculated by applying a
series of successive aggregations starting with the (more detailed) variables
and ending with the production of the index.
Owing to the lack of availability of data covering all the components for
each country, and in view of the need to maintain an adequate level of
accuracy, the composite index was calculated in a bottom-up pattern, where
the upper level index is calculated only where at least two thirds of its
components are available. This applies to all knowledge sub-indices and for
all countries. In cases where data for a variable was not available for at least
half of the countries, this variable was excluded from the calculation of the
overall composite indicators (i.e. excluded from the index structure).

The arithmetic aggregation formula was used to calculate all composite
indicators of the Index. The composite indicator (CI) is calculated by
aggregating its sub-components (SCj) as:

= ∑ ×

=1

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