25 The sectoral mix of projects in the 2003 census and 2008 survey are not directly comparable. Whereas the 2003 sample was dominated by hospitals, prisons and roads, the 2008 survey is predominantly accounted for by hospitals and groups of schools (Figure 1).
26 The 2008 survey results showed that 84 per cent of schools projects were completed on time or early. September term dates were thought by interview participants to be an additional incentive in completing to time. One project manager commented "We have not had a late school project, time is of the essence, we want a school for September not Christmas. So that's a strong message".
27 In addition to the 16 per cent of schools delivered late, 17 per cent of the 'on time' schools projects experienced late phases. This is important as these delays relate to the delivery of individual schools, which were part of a grouped project. In each case, however, the group of schools and hence the contract was delivered on time. There were no schools in the 2003 population for a comparison.
28 Fifty three per cent of hospitals reported completing to contracted time in 2008, compared to 82 per cent in 2003. Whereas schools have very set timetables, hospitals generally do not, which partially explains the sectoral difference in performance.