National survey of bait and berley use by recreational fishers

n/a. As discussed in Section 1.3, this is not to suggest that no such occurrence exists in the population overall – rather, that none was reported within the detection limits of the survey sample. Zero estimates should therefore be interpreted as ‘nil or negligible’.
Also, for estimates based on very small sub-samples (less than 5 respondents), the RSE is shown as n/a. Below this level, the influences of differing stratum weights can result in misleading error estimation. Although these survey estimates invariably refer to quite small proportions of the population or quantity estimates, where necessary, it is recommended that a 50% RSE be generally applied for conservative analysis purposes. On this basis, 95% confidence limits would approximate (+/-) 100% (see discussion below). If required, even higher upper limits may of course be applied.
To calculate 95% confidence intervals for a given survey estimate, the RSE is multiplied by 1.96 then applied to the estimate. For example, in Summary Table A (below), the estimate for Prawns/shrimp ‘Sold as Bait’ is 1,428,944 fishers and the RSE is 4.1%. The 95% confidence interval for this estimate is 1,314,114 – 1,543,774 with the lower limit calculated by 1,428,944 x 0.91964 (i.e. 100% - [1.96 x 4.1%]) and the upper limit by 1,428,944 x 1.08036 (i.e. 100% + [1.96 x 4.1%]).