5 We explored two possible causes of variations in the number of developed bids received by procuring authorities. First, we looked at whether the capital value of a deal was relevant, using an analysis of variance (ANOVA) test. Figure 17 shows that deals with different numbers of bidders did not vary significantly in capital value. Second, we tested for a relationship between geographical location and the number of developed bids received using a chi-squared test, following indications in interviews that this could be important. However, the test failed to confirm a statistically significant relationship (Figure 18).
17 | An ANOVA test of the relationship between capital value and the number of developed bids received | |||||
|
| Anova |
|
|
| |
| Sum of Squares | df | Mean Square | F | Sig. | |
Between groups | 20,921,043,489,022,000 | 1 | 20,921,043,489,022,450 | 0.193 | 0.662 | |
Within groups | 4,325,472,172,028,323,000 | 40 | 108,136,804,300,708,100 |
|
| |
Total | 4,346,393,215,517,345,000 | 41 |
|
|
| |
Source: National Audit Office |
|
|
|
| ||
18 | A chi-squared test of the relationship between geographical location and number of developed bids received by procuring authorities | ||||||||
|
| Strength of competition (number of bidders at ITN) | Total |
| Value | df | Asymp. Sig. (2-sided) | ||
|
| Two or fewer of ITN | Three or more at ITN |
| Pearson Chi-Square | 9.0311 | 6 | 0.172 | |
Region | London | 1 | 6 | 7 | Likelihood Ratio | 9.488 | 6 | 0.148 | |
| North West | 1 | 3 | 4 | Linear-by-Linear Association | 2.665 | 1 | 0.103 | |
| North East | 1 | 9 | 10 | N of Valid Cases | 44 |
|
| |
| Midlands | 6 | 4 | 10 |
|
|
|
| |
| South West | 2 | 2 | 4 |
|
|
|
| |
| South/South East | 3 | 2 | 5 |
|
|
|
| |
| National | 1 | 3 | 4 |
|
|
|
| |
Total |
| 15 | 29 | 44 |
|
|
|
| |
Source: National Audit Office |
|
|
|
|
|
| |||
Note 1 Twelve cells (85.7 per cent) have expected count less than five. The minimum expected count is 1.36. | |||||||||