• Tidak ada hasil yang ditemukan

Factor Analysis and Confirmatory Factor Analysis for Brand Equity

172 used to measure Empathy, were relevant. Therefore, Empathy is a key underlying factor or dimension of Service Quality in this study.

The Factor analysis and Confirmatory Factor analysis for the Brand Equity construct is discussed in the next section.

173 Table 5.29: Total Variance Explained for Brand Equity

Factor

Initial Eigenvalues

Extraction Sums of Squared Loadings

Rotation Sums of Squared Loadings

Total

% of Variance

Cumulative

% Total

% of Variance

Cumulative

% Total

% of Variance

Cumulative

%

1 10.907 43.626 43.626 10.479 41.916 41.916 4.907 19.630 19.630

2 1.725 6.902 50.528 1.266 5.065 46.982 3.105 12.419 32.049

3 1.329 5.315 55.843 .871 3.483 50.464 2.608 10.431 42.480

4 1.117 4.468 60.311 .610 2.439 52.903 2.606 10.423 52.903

Extraction Method: Principal Axis Factoring.

It is evident from the rotated factor matrix (Table 5.30), that Factor 1 loaded heavily on ‘Loyalty’

and more strongly on ‘Key Associations and Differentiation’ items/variables. Factor 2 loaded strongly on the ‘Loyalty’ items/variables. Factor 3 loaded on the ‘Awareness’ items/variables and Factor 4 loaded on ‘Perceived Quality’ items/variables. Based on these loadings, Factor 1 is called

‘Key Associations and Differentiation with Loyalty’, Factor 2 is called ‘Loyalty’, Factor 3 is called

‘Awareness’ and Factor 4 is named ‘High Quality and Reliability”.

174 Table 5.30: Rotated Factor Matrix Service Quality

Brand Equity Variables Factor

1 2 3 4

C1 High Quality Brand .623

C2 High Reliability .642

C3 Very Good Quality Brand .442

C4 Is a Leading Brand

C5 Growing in Popularity .449

C6 Innovative and Very Advanced .476

C7 More Aware of the Brand than Other Brands .672

C8 Easily Recognizable Brand .679

C9 Characteristics of Institution Easily Come to Mind .639

C10 I know what the Brand Looks Like .446

C11 Will not Study at another Institution .690

C12 Is My First Study Choice .648

C13 I am Loyal to Institution .565

C14 Would Recommend this Institution .567 C15 Would Not Switch to Another Institution .720

175

C16 Will Study at this Institution Despite Fee Increases .640

C17 Institution Provides Good Value for Money .435 C18 I have Good Reason to Support this Institution .516

C19 Institution Has a Good Personality .633 .441 C20 Institution's Brand is Interesting .535

C21 I Have a Positive Image of People Studying Here .648 C22 I Trust this Institution's Brand .741 C23 I Admire this Institution's Brand .749

C24 The Institution Has Credibility .608 .434 C25 The Institution's Brand is Better than that of Others .430 .447

Extraction Method: Principal Axis Factoring.

Rotation Method: Varimax with Kaiser Normalization.

Based on the factor analysis and its underlying dimensions, a Confirmatory Factor Analysis (CFA) using AMOS version 23 was conducted for the brand equity construct, which is based on the Aaker’s (1996) Model. The CFA revealed a three-factor brand equity model as depicted in Figure 5.8.

176 Figure 5.8: Measurement Model for Brand Equity

A reliability analysis (Table 5.31) revealed the Cronbach Alpha scores for each dimension was above 0.7, and were deemed reliable (Andrew et al., 2011:202).

177 Table 5.31: Reliability Scores for Brand Equity Construct Dimensions Confirmed by CFA

Variable Cronbach’s Alpha

Key Associations and Differentiation (KASSO) 0.842

Loyalty (LOY) 0.711

Awareness (AWARE) 0.813

The figure represents the Brand Equity dimensions through conducting a CFA where KASSO=Key Associations and Differentiation, LOY=Loyalty and AWARE=Awareness.

The model fit indices for the Brand Equity model appear Table 5.32 below

Table 5.32: Model Fit Indices for Aaker’s (1996) Brand Equity Model (Brand Equity Construct)

Measure Threshold Indices for Model Comment

Chi-square/df (cmin/df) <3 good, <5 sometimes allowed (Hu & Bentler,

1999). 1.652

ACCEPTABLE

p- value >0.05 (Hu & Bentler). 0.023 NOT ACCEPTABLE

CFI >0.9 (Hu & Bentler, 1999).

0.989 ACCEPTABLE

GFI 0.9 minimum (Hu &

Bentler, 1999).

0.978 ACCEPTABLE

AGFI Equal to or >0.9 (Hooper et al., 2008 cited in Kats, 2013:103).

0.959 ACCEPTABLE

178

Measure Threshold Indices for Model Comment

NFI >0.9 (Bentler, 1995). 0.973 ACCEPTABLE

RMSEA <0.06 (Hu & Bentler, 1999).

0.040 ACCEPTABLE

PCLOSE >0.05 (Hu & Bentler, 1999.

0.743 ACCEPTABLE

Although the p-value of the model is not in accordance with the recommended threshold, based on the values of the other fit indices, which are in accordance with the recommended thresholds, it is inferred that the model is a good fit.

To check whether the factors identified in the Brand Equity model display convergent and discriminant validity, appropriate analyses were conducted shown in Table 5.33. According to Esposito (2010:696) convergent validity exists when AVE is greater than 0.5. In addition, when MSV is less than AVE and ASV is less than AVE, discriminant validity can be claimed (Hair et al., 2009 cited in Ernst, 2015:38). The table below was generated by a template put together by Professor Gaskin to test for convergent and discriminant validity in confirmatory factor analysis (Statwiki, n.d.). From table, AVE values for each factor in the model is greater than 0.5 and hence convergent validity can be claimed for each factor (Esposito, 2010:696). Furthermore, for each factor, the MSV and ASV values are less than AVE and hence discriminant validity can be claimed (Hair et al., 2009 cited in Ernst, 2015:38).

Table 5.33: Convergent and Discriminant Validity Indices

CR AVE MSV ASV KASSO loy aw

KASSO 0.844 0.521 0.498 0.445 0.722

loy 0.720 0.566 0.392 0.305 0.626 0.752

aw 0.817 0.692 0.498 0.358 0.706 0.467 0.832

179 Please note, for clarification of terms pertaining to the table, KASSO=Key Associations and Differentiation, loy=Loyalty, and aw=Awareness.

The data in the model was tested for normality (Table 5.34). According to Kline (2005) cited in Harrington (2009), it is suggested that variables having absolute values of greater than 3 for the skew index and absolute values of greater than 10 for the kurtosis index indicate normality problems. The skewness and kurtosis values in the dataset for each variable is well within the specified range and hence the assumption of normality is met.

Table 5.34: Assessment of Normality

Variable min max skew kurtosis C23_1 1.000 7.000 -.967 .519 C7_1 1.000 7.000 -.880 .006 C8_1 1.000 7.000 -1.149 .616 C15_1 1.000 7.000 .018 -1.317 C16_1 1.000 7.000 .522 -1.041 C6_1 1.000 7.000 -.779 .021 C14_1 1.000 7.000 -1.122 .312 C18_1 1.000 7.000 -.662 -.190 C20_1 1.000 7.000 -.733 .033

Multivariate 36.406

The analysis into Brand Equity reveals that for this study, the Brand Equity construct has three dimensions that are Key Associations and Differentiation, Loyalty, and Awareness. The dimension Perceived Quality did not load into the model. Furthermore, of the 25 variables used to measure

180 Brand Equity, only nine applied to the revised model. This suggests that Aaker’s (1996) model of Brand Equity may not be totally applicable to the sample of students studied.

The Factor analyses and Confirmatory Factor Analyses pertain to objective three of the study, which is concerned with the explanatory factors for both Service Quality and Brand Equity in the study. Interestingly, the dimensions of both Service Quality and Brand Equity have not loaded exactly as the original conceptual/theoretical models that these constructs were based on. This could suggest that different Service Quality and Brand Equity models and measurements may be applicable to the South African higher education context.