separate Cloud services will they provide information concerning respondents' preferences. Taking the exponent to the total utility of Cloud A and dividing it by the exponent of the total utility of Cloud A and Cloud B, provides the probability that respondents will chose Cloud A when given the choice between Cloud A and Cloud B. This is represented in the following formula: per year, has a CPA checking the reliability, and makes the information publicly available (Total Utility = -2.09) would give: P(A|A&B) = exp(-1.17) / (exp(-1.17) + exp( -2.09)) P(A|A&B) = 71.5% and P (A| A & B) = exp(TU Cloud A) / (exp(TU Cloud A) + exp(TU Cloud B) For example, if we assume that a respondent had to choose between the previously described Cloud A and Cloud B, which provides low transparency, costs 1 euro P(B|A&B) = exp(-2.09) / (exp(-1.17) + exp(-2.09)) P(B|A&B) = 28.5% This shows that 71.5% of respondents would have preferred Cloud A over Cloud B. In Table 2 we have Table 2. Choice Tasks Used Choice set Transparency Costs Reliability Audience 1 1 2 2 3 3 4 4 5 5 6 6 7 7 8 8 9 9 10 10 11 11 12 12 C1 C1 C2 C2 low Medium Medium High High Medium low High High Medium low High Medium High low Medium low Medium Medium low low Medium Medium High Medium High Medium low 60 30 30 0 30 60 0 60 30 0 30 60 60 30 0 30 0 60 0 60 30 0 30 60 30 0 0 60 CPa CP Ia CP CP Ia CPa Ia CPa Ia Ia CPa CPa CP CP CPa CPa CP CPa Ia Ia CP CP Ia CP Ia Ia CPa Only you Publicly Only you Publicly Only you Publicly Only you Only you Only you Publicly Only you Publicly Only you Publicly Only you Publicly Publicly Only you Publicly Only you Publicly Only you Only you Publicly Only you Only you Only you Publicly Predicted Actual Probability Probability 50.00% 50.00% 73.11% 26.89% 73.11% 26.89% 26.89% 73.11% 50.00% 50.00% 73.11% 26.89% 50.00% 50.00% 50.00% 50.00% 73.11% 26.89% 73.11% 26.89% 26.89% 73.11% 26.89% 73.11% 1.80% 98.20% 99.33% 0.67% 10.07% 89.93% 20.26% 79.74% 88.39% 11.61% 72.31% 27.69% 26.11% 73.89% 67.92% 32.08% 13.59% 86.41% 72.11% 27.89% 73.89% 26.11% 95.85% 4.15% 8.55% 91.45% 69.21% 30.79% 7.45% 92.55% 99.15% 0.85% Note. Predicted probability concerns probability respondents' choice if the utility of all attributes and levels are assumed equal. actual probability concerns probability respondents' choice calculated from obtained utilities from the survey. C1 and C2 are control tasks. December 2015 ∕ IEEE Technology and Society Magazine 79