You are on page 1of 2

1

4.3.3.1 The Total Variance Explained of the Retained Factors

The EFA test was run multiple times to obtain a clean pattern matrix, and variables that had

cross-loadings were dropped. However, it is required that the retained variables should be able to

explain at least 50% of the total variance (Streiner, 1994). Hair et al. (2014) indicated that a

cumulative variance of 60% or higher is commonly used to justify the fit of the retained factors.

For this study, the retained variables presented a cumulative variance of 72%. The cumulative

variance, as revealed in the test, is as per table 4.10

Table 4. 1: Total Variance Explained


Factor Extraction Sums of Squared Loadings Rotation Sums of Squared Source:
Loadings
Total % of Variance Cumulative % Total
1 13.539 41.028 41.028 10.243
2 2.257 6.839 47.867 8.210
3 1.675 5.076 52.942 9.628
4 2.030 6.151 59.094 7.737
5 1.100 3.333 62.426 10.170
6 .987 2.991 65.417 3.003
7 .842 2.550 67.968 8.668
8 .853 2.584 70.552 10.616
9 .632 1.916 72.467 8.233
Researcher (2019)

4.3.3.2 EFA

Pattern Matrix and Reliability Assessment

A clean pattern matrix was obtained, and all loadings were > 0.3. Guadagnoli & Velicer (1988)

explained that low loadings in EFA are acceptable if the sample size is ≥ 300. As a guideline for

the relationship between sample size and factor loading, it is indicated by Hair et al. (2014) that,
2

for a sample size ≥ 350, the loadings that ≥ 0.30 are acceptable. Costello & Osborne (2005)

argued that after rotation factors that have at least three items with no or fewer cross-loadings

and loadings of above 0.30 fits best to the data. This study had a sample size of 403, and the

lowest loading was 0. 346. As indicators for each construct were loading in their respective cells,

of hypotheses) was followed. The structural model is as presented in Figure 4.3.

You might also like