The original SDD index does not assign fixed weights but uses the Principal Component Analysis (PCA) to reveal the weights. This calculation was done through the statistical software package S-Plus.
Weights for the CT and DA level using Principal Component Analysis
|
Factor |
Weighted Score for the CT |
Weighted score for the DA |
|
Unemployment |
0.3508682 |
0.2596363 |
|
Youth Unemployment |
0.1992592 |
0.1834997 |
|
Overcrowding |
0.137622 |
0.1480046 |
|
Lone Parents |
0.1178725 |
0.1294785 |
|
Unskilled Workers |
0.07794345 |
0.1167971 |
|
Elderly Households |
0.06276595 |
0.09068184 |
|
Lacking Amenities |
0.05366872 |
0.07190312 |
|
Final PCA score |
1.00000002 |
1.00000116 |
These weights were then assigned to the factors and the seven weighted factors were then added to the overall SDD-Index.
The problems associated with the weighting using a PCA is values (weights) are assigned based on the correlation between all variables. While this method may be seen as incorporating objectivity into the analysis the PCA does not take into account any logical decisions based on experience or human society structure. As well, the weights for each factor will change if the order of factors is changed. The incorporation of an expert group model would add some validity to this method to determine if the PCA overestimated or underestimated the importance of each factor.