CHAPTER 4 DATA ANALYSIS AND PRESENTATION OF RESULTS
4.4 Exploratory Factor Analysis
47
Figure 4.5: Previous Respondents’ Industry of Employment
48
Item Removed Removed in
Cost cutting measures have promoted creativity in a manner in which service is delivered
I am forced to review my yearly targets because of the cost cutting measures
Cost cutting measures have no effect on the staff morale Cost cutting measures should have been implemented differently
I am able to achieve my yearly targets regardless of the cost cutting measures
My clients are still happy with the quality of work that I provide since the implementation of cost cutting measures
Round Two New platforms for creativity and innovation are there in my
organisation in the midst of cost cutting
Cancellation of some of the work related trips due to cost cutting has led to a low morale on staff
I am not satisfied with my work under the implementation of cost cutting measures.
I dislike cost cutting and I believe that they should not be implemented in our organization
A total of nineteen items were retained and were subjected to a final round of EFA using the Direct Oblimin rotation. The questionnaire used was developed by the researcher and had to undergo some diagnostic tests. This was achieved through the application of the Kaiser-Meyer-Olkin (KMO) criterion.
Table 4.2: Kaiser-Meyer-Olkin (KMO) and Bartlett's Test of Sphericity KMO and Bartlett's Test
Kaiser-Meyer-Olkin Measure of Sampling Adequacy. .838
Bartlett's Test of Sphericity Approx. Chi-Square 725.921
df 171
Sig. .000
Based on the Table 4.2 above, the KMO was 0.838 which is above a threshold of 0.700 and the Bartlett's Test of Sphericity – significance (sig) value is 0.000 which is
49
below 0.01 (Field, 2013). This allowed the researcher to conclude that the properties of the correlation matrix of the item scores were appropriate for factor analysis.
4.4.1 Total Variance Explained (Eigenvalues and the Number of Factors Problem)
The purpose of the researcher was to retain only the main components for analysis.
In this study this was achieved through the application of The Kaiser-Meyer-Olkin (KMO) criterion. The researcher had to determine how many factors to retain for subsequent analyses. This criterion was suggested by Kaiser (1960), the idea was to retain only factors with eigenvalues larger than 1.000. In principle this is means that, unless a factor extracts at least as much as the equivalent of one original variable, then it can be released. Table 4.3 below shows the factor loadings of the retained factors.
Table 4.3: Total Variance Explained
Total Variance Explained
Component
Initial Eigenvalues Extraction Sums of Squared Loadings
Rotation Sums of Squared Loadingsa Total % of
Variance
Cumu-
lative % Total % of Variance
Cumu-
lative % Total 1 6.934 36.494 36.494 6.934 36.494 36.494 5.496 2 1.685 8.866 45.360 1.685 8.866 45.360 3.658 3 1.634 8.603 53.962 1.634 8.603 53.962 3.977 4 1.211 6.373 60.335 1.211 6.373 60.335 1.492 5 .923 4.860 65.196
6 .829 4.365 69.560 7 .758 3.990 73.550 8 .716 3.770 77.320 9 .625 3.292 80.611 10 .593 3.123 83.735 11 .524 2.758 86.492 12 .479 2.522 89.015 13 .461 2.428 91.443 14 .387 2.035 93.478 15 .337 1.775 95.253 16 .280 1.476 96.729 17 .258 1.357 98.087
50
Total Variance Explained
Component
Initial Eigenvalues Extraction Sums of Squared Loadings
Rotation Sums of Squared Loadingsa Total % of
Variance
Cumu-
lative % Total % of Variance
Cumu-
lative % Total 18 .209 1.101 99.188
19 .154 .812 100.000
Extraction Method: Principal Component Analysis.
a. When components are correlated, sums of squared loadings cannot be added to obtain a total variance.
Any component with a total above 1.000 in the table above was retained. The first four factors above have total values over 1.000 and cumulatively explain 60.335 percent of data which is the majority. Since the above table alone provides limited data, the use of Scree Plot graph below was considered. The Scree Plot below shows the first break after the first component meaning that it explains a lot more variance than the rest of the components. After the fourth component there, however, is another break which suggests that 4 components may be appropriate (refer to Figure 4.6).
Figure 4.6: Scree Plot
After the above assessment criteria and based on the strongest loading items, the four factors were then labelled as: Cost cutting relationship to service quality, Cost cutting relationship to staff morale, Cost cutting relationship to job satisfaction and Consistency in cost cutting implementation.
51 4.4.2 Communalities
The communalities table assists to demonstrate the relationship between the variances (factor aspects) such as the correlation between one cost cutting measure aspect to other aspects. Extraction communalities are estimates of the variance in each variable accounted for by all factors in the factor solution. This is accomplished through undertaking Kaiser Normalization criteria using SPSS technique. It produces extraction with respect to each facet. Small values below 0.300 indicate variables that do not fit well with the factor solution, and should possibly be dropped from the analysis. Based on the Table 4.4 below all values indicate extraction values above the 0.300 threshold. High values indicate variables that fit well with the factor solution.
Table 4.4: Communalities
Communalities
Item Initial Extraction
Cost cutting measures reduce quality of services provided to clients
1.000 .579
Cost cutting measures have minor to no effect on the quality of work delivered to clients
1.000 .518
I am able to achieve my yearly targets regardless of the cost cutting measures
1.000 .609
My colleagues and I have resorted to pooling resources to achieve our yearly targets
1.000 .692
I do the bare minimum since the implementation of the cost cutting measures towards my duties
1.000 .698
My clients are still happy with the quality of work that I provide since the implementation of cost cutting measures
1.000 .488
Cost cutting measure provides no room for creativity as we have to avoid costly actions in our organisation
1.000 .476
Creativity manifests from cost cutting measures thus improves the staff morale
1.000 .531
Cost cutting measures frustrate me 1.000 .637
52
Communalities
Item Initial Extraction
I look forward to coming to work since the implementation of cost cutting measures
1.000 .527
The cost cutting model implemented has improved the way we look at government spending
1.000 .460
I am satisfied with my work under the implementation of cost cutting measures
1.000 .694
I would recommend my department/ unit to external parties regardless of the cost cutting measures
1.000 .626
I like cost cutting and I believe that all organization should implement them
1.000 .650
Cost cutting measures are implemented across the department consistently
1.000 .717
Communication relating to the cost cutting is adequate 1.000 .616 I understand why it is important for the organization to
implement cost cutting measures
1.000 .676
I encourage my colleagues to comply with cost cutting measures
1.000 .664
I have noted significant improvements since the introduction of cost cutting measures in government spending
1.000 .605
Extraction Method: Principal Component Analysis.
Item “Cost cutting measures are implemented across the department consistently” has the highest (71.7%) communality or shared relationship with other variables as compared to “The cost cutting model implemented has improved the way we look at government spending” item which has the lowest (46.0%) communality.