This content requires Adobe Flash Player version
or later.
Either you do not have Adobe Flash Player installed,
or your version is too old,
or there is a problem with your Flash installation and we were unable to detect it.
Correlation analysis is one of the most widely used and reported statistical methods in summarising quantitative research data (Williams, 2007: 67). Variables that are not associated with at least some of the other variables would not contribute to the analysis. Those are variables that were identified as having low correlations of less than a threshold value of 0.5 and were eliminated from further analysis.
TABLE 5: CORRELATION MATRIX OF THE VARIABLES Mean
WE EC RE ET
SS PL
4.662 4.869 4.470 4.972
4.567 4.945 Std dev
1.089 1.135 1.144 0.997
1.296 1.186 WE
1.000 0.734 0.738 0.757
0.622 0.644 EC
0.734 1.000 0.792 0.686
0.534 0.505 RE
0.738 0.792 1.000 0.796
0.617 0.677 ET
0.757 0.686 0.796 1.000
0.646 0.701 SS
0.622 0.534 0.617 0.646
1.000 0.657 PL
0.644 0.505 0.677 0.701
0.657 1.000
Table 4 indicates that, among other, working environment is positively correlated to skills shortage with a coefficient of 0.622. Table 4 further shows that skills shortage is positively correlated to employment conditions with a coefficient of 0.534 and is also positively correlated to resources with a coefficient of 0.617. Education and training is positively related to skills shortages with a coefficient of 0.646. Table 4 indicates that propensity to leave and skills shortages have a significant positive correlation with a coefficient of 0.657.
Regression analysis
Table 6 indicates the regression analysis results of the influence of working environment, employment conditions, resources and education and training on skills shortages in gold mines.
TABLE 6: REGRESSION ANALYSIS: INFLUENCE OF WORKING ENVIRONMENT, EMPLOYMENT CONDITIONS, RESOURCES AND EDUCATION AND TRAINING ON SKILLS SHORTAGES
REGRESSION SUMMARY FOR DEPENDENT VARIABLE: SKILLS SHORTAGES Parameter
Working environment (WE) Employment conditions (EC) Resources (RE)
Education and training (ET) R2
R 69%
* = p < 0.05 ** = p < 0.01 *** = p < 0.001
0.4703 B
0.311 -0.034 0.228 0.399
F 65.484
Std Error 0.087
0.085 0.096 0.100
T-value 3.574
-0.397 2.379 3.993
Std Error of P estimate 0.9497 p<0 .00000
P-value 0.001***
0.691
0.017** 0.001***
IMPACT OF SKILLS SHORTAGES ON PROPENSITY TO LEAVE IN SOUTH AFRICAN GOLD MINES 731