Page 70 - Azerbaijan State University of Economics

THE JOURNAL OF ECONOMIC SCIENCES: THEORY AND PRACTICE, V.75, # 2, 2018, pp. 64-78

value of the brand to the consumers, being an integral component of brand equity
(Malik, Naeem & Munawar 2012). Questions related to brand image were taken from
main study measurements of Heinberg, Ozkaya & Taube (2017), which is based on
Keller’s Brand image 1993 and brand was replaced with university.

The majority of the students in this sample, went to United States or countries in
Europe to earn either a Bachelor or Master degree. Age of the participants was in the
range of 18 to 30 years. The regions that were included in this study were East &
South Asia & Pacific (28.9%), Europe and Central Asia (49.7%), South and North
America (14.5%), Middle East and North Africa (6.9%). Students from the countries
in Europe, America and Asia answered Survey questions. Some countries included in
the sample were; US, Canada, Brazil, Ecuador, Turkey, Russia, Greece, Macedonia,
Bulgaria, Netherlands, Sweden, Lebanon, Iran and, Morocco.

5.METHODOLOGY

The analysis was done in two phases first a principal component analysis, second a
simple regression. The principal component analysis was used to group together the
survey questions with little variance and discover the factors from the survey
questions. Once the factors were discovered, a simple regression was run to explore
the strength of the factors on enrollment intention.

Principal component analysis

Based on scree plot, two factors were extracted from the data. In the extraction method
principal component analysis, the first eleven components had eigenvalues greater
than “1” which was around 15 components. There was a huge drop between
eigenvalue of component 1 and component 2. The first two components explained
60.77% and 13.46% of total variance in order and third component explained 5.49%.
The first two components yielded as 74.24% as the cumulative value. Eigenvalues of
first, second and third components were 9.12, 2.02 and 0.82 in order, therefore, the
first two components were retained for the study.

After rotation, in rotation sums of squared loadings, first two components together
equaled to ((5.72 + 5.42)/15) = 74 % of the total variance. In rotated component
matrix, in factor loadings, it is found that university created communication (UCC)
and user generated content (UGC) with overall eight independent variables loaded to
one factor (Table 1). UCC and UGC came together as one factor, which we called
them Co-communication (collaborative communication). Brand image (BI) and brand

70