4.3 Benthic Power Analysis 4.3.1 East Anglia THREE and FOUR.
Table 4.3 presents the Power inherent from the 2010 surveys (MESL 2011) for benthic surveys from East Anglia THREE and FOUR.
Table 4.3. Results from power analysis for East Anglia THREE and FOUR. Effect Size
0.2 0.5 0.8
EA FOUR (n = 49) EA THREE (n = 48) Power H' Power S Power H' Power S 0.82 0.99 1
1 1 1
0.67 0.99 1
0.99 1 1
The results show that only one test, for Shannon (H’) at an effect size of 0.2 (East Anglia THREE), fails to reach the required Power level (0.8). At an es of 0.81 = 15% detectable change (Table 4.2) for the Shannon metric in East Anglia THREE however, the 0.8 level (effect size) is more than adequate to detect change at less than the required 50% level. This is true for all benthic surveys from East Anglia THREE and FOUR and therefore we can state that the sample numbers are sufficient, and that an es of 0.8 as a benchmark is also suitable for determining whether the number of sites sampled in 2010 was suitable to characterise the site and allow a future detection of change.
1 Please note; 0.8 is the critical threshold for statistical Power, and care should be taken to avoid confusion between this value and the effect size of 0.8
