5. Outcomes
5.3 Expert panel survey outcome
The expert panel survey was a more detailed multi-disciplinary comparison of the two development paths. The comparison followed the structure of the criteria set as describe in Figure 16 with four of the five main criteria groups being accommodated in the expert panel survey. The fifth main criteria group (public acceptance) was measured by the public survey and was only accounted for in the aggregation process as summarised in Table 17, Table 18 and Table 19.
Two issues were of importance for the expert panel survey: first, the level of consensus among experts regarding the scores and criteria weights (relative importance of the differentiating criteria) and second, the aggregated weighted total score for each development path. The level of consensus was important since it impacted on the decision regarding which point parameter in each distribution will be used in the aggregation process. The aggregation process per se will allow decision-makers to make water resource allocation decisions with greater confidence without harbouring the fear that these decisions will harm their political positions.
94 Normal probability plots (not shown) indicated that a significant number of criteria were non-normally distributed with significant variation evident for both scores and weights (see Figure 17).
This implies a relatively low level of consensus among experts regarding the level of importance of the different criteria. However, differences were expected since the panel was composed of experts from different fields of interest.
Box & Whisker Plot (Scores)
Median 25%-75%
Min-Max PubopinA ExpopinA WBYIELDA WBTIMEA FINURVA FINCTCEA FINTARFA ENIFRA ENWASTEA ENGWRECA ENFOODA ENLOSSA SEDEPRAINA SEVOLA SEAGRIA SEMULTA SEURBA SERECTA
-20 0 20 40 60 80 100
120 Box & Whisker Plot (Criteria Weights)
Median 25%-75%
Min-Max WaterB Fin Environ SocioEcon PubAcc ConfYield Timing URV ConfCostEst Tariffs IFR WasteDisp Recharge FloodContr BioDiv DepRain VolRe-alloc AgriEmpl Multipl Urban RecTour
-0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
Figure 17: Box and whisker plot for scores and weights of all criteria Source: (De Lange, 2005)
A single parameter representing each distribution of scores and weights was needed for aggregation purposes (therefore assuming normality). For the sake of comparison, the mean, mode and median were used in three separate aggregation runs for comparative calculations. Answers of the same magnitude were obtained with all three indicators displaying a majority for Development Path B.
The study settled on presenting only the mean for the weight structure, which gave the most balanced view of the criteria for display purposes (see Figure 18 for the mean weight of each of the five main criteria groups and Figure 19 for the average of the different sub-criteria).
95 Relative importance: Main criteria
0%
5%
10%
15%
20%
25%
30%
Water balance Financial aspects Socio-economic aspects
Environmental aspects
Public acceptance
Weight (%)
Figure 18: Mean weights: main criteria groups Source: (De Lange, 2005)
Relative importance: Sub-criteria
11.91%
10.15%
9.08%
7.46% 7.70%
3.93%
2.33%
5.27%
2.53%
1.70% 2.07%
8.57%
4.46% 4.35%
2.74%
6.36%
0%
2%
4%
6%
8%
10%
12%
14%
Conf idence in yield
Timing (ability to supply the
demand)
URV Conf idence in total cost estimates
Tar if f changes necessar y to maintain ser vice
Dependency on natur al r ainf all
Volume of water allocated f r om r ur al to ur ban ar eas
Impacts on agr icultur al pr oduction and
empl oyment
Multi plier s Ur bani sation Recr eati on and tour ism
Expected impacts on IFR
Waste disposal and dil uti on
ef f ect Gr ound water
r echar ge.
Fl ood and er osi on contr ol
Loss of biodiver sity
Weight (%)
Figure 19: Mean weights: sub-criteria groups Source: (De Lange, 2005)
Aggregated total scores for the two development paths were obtained by aggregating the weighted scores obtained from each sub-criterion. A weighted score for each sub-criterion was obtained by multiplying mean (mode or median) scores by the mean (mode or median) weight of the particular sub-criterion. Total scores for development paths A and B were, therefore, compared in terms of the mean (see Table 17), the mode (see Table 18) and the median (see Table 19). Answers of the same magnitude were obtained and using the mean was settled on, which gave the most balanced
96 view of the criteria. (Note the placing of the outcome of the public survey under the public acceptance criteria group.)
Table 17: Aggregated score for development paths A and B (mean)
Main criteria Sub-criteria: Weight Cumulative weight A B A B
Water balance 1.000 0.221
Confidence in yield 0.529 0.119 0.387 0.613 0.046 0.073
Timing (ability to supply the demand) 0.471 0.101 0.370 0.630 0.038 0.064
Financial aspects 1.000 0.242
URV 0.389 0.091 0.682 0.318 0.062 0.029
Confidence in total cost estimates 0.304 0.075 0.466 0.534 0.035 0.040
Tariff changes necessary to maintain service 0.307 0.077 0.631 0.369 0.049 0.028
Socio-economic aspects 1.000 0.178
Dependency on natural rainfall 0.221 0.039 0.153 0.847 0.006 0.033
Volume of water allocated from rural to urban areas 0.130 0.023 0.267 0.733 0.006 0.017
Impacts on agricultural production and employment 0.296 0.053 0.382 0.618 0.020 0.033
Multipliers 0.125 0.025 0.406 0.594 0.010 0.015
Urbanisation 0.115 0.017 0.414 0.586 0.007 0.010
Recreation and tourism 0.112 0.021 0.445 0.555 0.009 0.011
Environmental aspects 1.000 0.265
Expected impacts on IFR 0.311 0.086 0.270 0.730 0.023 0.063
Waste disposal and dilution effect 0.174 0.045 0.342 0.658 0.015 0.029
Ground water recharge. 0.175 0.044 0.409 0.591 0.018 0.026
Flood and erosion control 0.100 0.027 0.544 0.456 0.015 0.012
Loss of biodiversity 0.241 0.064 0.359 0.641 0.023 0.041
Public acceptance 1.000 0.094 0.386 0.614 0.036 0.058
1.000 0.418 0.582
Score (unweighted) Score (weighted)
Source: (De Lange, 2005)
Table 18: Aggregated score for development paths A and B (mode)
Main criteria Sub-criteria: Weight Cumulative weight A B A B
Water balance 1.000 0.330
Confidence in yield 0.700 0.210 0.400 0.600 0.084 0.126
Timing (ability to supply the demand) 0.300 0.120 0.500 0.500 0.060 0.060
Financial aspects 0.900 0.175
URV 0.300 0.090 0.700 0.300 0.063 0.027
Confidence in total cost estimates 0.200 0.045 0.400 0.600 0.018 0.027
Tariff changes necessary to maintain service 0.400 0.040 0.700 0.300 0.028 0.012
Socio-economic aspects 0.900 0.165
Dependency on natural rainfall 0.200 0.030 0.000 1.000 0.000 0.030
Volume of water allocated from rural to urban areas 0.150 0.030 0.000 1.000 0.000 0.030
Impacts on agricultural production and employment 0.300 0.045 0.500 0.500 0.023 0.023
Multipliers 0.100 0.030 0.500 0.500 0.015 0.015
Urbanisation 0.100 0.015 0.500 0.500 0.008 0.008
Recreation and tourism 0.050 0.015 0.500 0.500 0.008 0.008
Environmental aspects 1.050 0.190
Expected impacts on IFR 0.400 0.050 0.200 0.800 0.010 0.040
Waste disposal and dilution effect 0.200 0.025 0.400 0.600 0.010 0.015
Ground water recharge. 0.200 0.050 0.500 0.500 0.025 0.025
Flood and erosion control 0.050 0.015 0.400 0.600 0.006 0.009
Loss of biodiversity 0.200 0.050 0.400 0.600 0.020 0.030
Public acceptance 1.000 0.100 0.386 0.614 0.039 0.061
0.960 0.415 0.545
Score (unweighted) Score (weighted)
Source: (De Lange, 2005)
97 Table 19: Aggregated score for development paths A and B (median)
Main criteria Sub-criteria: Weight Cumulative weight A B A B
Water balance 1.000 0.170
Confidence in yield 0.500 0.080 0.400 0.600 0.032 0.048
Timing (ability to supply the demand) 0.500 0.090 0.357 0.643 0.032 0.058
Financial aspects 0.960 0.200
URV 0.330 0.070 0.700 0.300 0.049 0.021
Confidence in total cost estimates 0.300 0.070 0.500 0.500 0.035 0.035
Tariff changes necessary to maintain service 0.330 0.060 0.700 0.300 0.042 0.018
Socio-economic aspects 0.943 0.157
Dependency on natural rainfall 0.200 0.040 0.100 0.900 0.004 0.036
Volume of water allocated from rural to urban areas 0.143 0.020 0.308 0.692 0.006 0.014
Impacts on agricultural production and employment 0.300 0.045 0.400 0.600 0.018 0.027
Multipliers 0.100 0.020 0.438 0.563 0.009 0.011
Urbanisation 0.100 0.018 0.462 0.538 0.008 0.009
Recreation and tourism 0.100 0.015 0.500 0.500 0.008 0.008
Environmental aspects 1.000 0.230
Expected impacts on IFR 0.300 0.063 0.286 0.714 0.018 0.045
Waste disposal and dilution effect 0.200 0.045 0.400 0.600 0.018 0.027
Ground water recharge. 0.200 0.050 0.400 0.600 0.020 0.030
Flood and erosion control 0.100 0.023 0.500 0.500 0.011 0.011
Loss of biodiversity 0.200 0.050 0.357 0.643 0.018 0.032
Public acceptance 1.000 0.100 0.386 0.614 0.039 0.061
0.857 0.366 0.491
Score (unweighted) Score (weighted)
Source: (De Lange, 2005)
The outcome suggests that Development Path B was the preferred option suggesting a willingness to pay for “greener” water and public acceptance of increased water tariffs to accommodate the implementation of such strategies. However, care must be taken not to use the outcome of this study as an “over-generalization” to promote “greener” allocation strategies in the CCT since, it should be borne in mind that an MCDM exercise merely aids in the decision-making process and does not replace the water manager as the final decision maker.