Skip to main content
Data Collection Framework - DCF

Precision level

Precision level DCF

Where reference is made to precision/confidence level the following distinction shall apply:

Level 0: 

Precision of parameter is less than for 'level 1' ie parameter cannot be estimated within precision of plus or minus 40 % for a 95 % confidence interval and the coefficient of variation (CV) is higher than 20 %

Level 1: 

Level making it possible to estimate a parameter either with a precision of plus or minus 40 % for a 95 % confidence level or a coefficient of variation (CV) of 20 % used as an approximation;

Level 2: 

Level making it possible to estimate a parameter either with a precision of plus or minus 25 % for a 95 % confidence level or a coefficient of variation (CV) of 12,5 % used as an approximation;

Level 3: 

Level making it possible to estimate a parameter either with a precision of plus or minus 5 % for a 95 % confidence level or a coefficient of variation (CV) of 2,5 % used as an approximation.

Level 4: 

Level making: exhaustive data collection.

NR 

Not Relevant. It is not possible to provide a meaningful calculation of the precision - eg. because the variance is too large.

NA 

Not Applicable. There is no obligation to estimate precision for this parameter.

Coefficient of variation (CV) is preferred when applicable (Accuracy indicator).

Precision level DCR

PRECISION

 

Description

0

=

precision of parameter is less than for level 1. ie parameter cannot be estimated within precision of plus or minus 25% with a 95% confidence interval

1

=

level 1: level making it possible to estimate a parameter with precision of plus or minus 25 % for a 95 % confidence level;

2

=

level 2: level making it possible to estimate a parameter with precision of plus or minus 10 % for a 95 % confidence level

3

=

level 3: level making it possible to estimate a parameter with precision of plus or minus 5 % for a 95 % confidence level.

4

=

exhaustive data collection

NR

=

(NOT RELEVANT) it is not possible to provide a meaningful calculation of the precision - eg. because the variance is too large

NA

=

(NOT APPLICABLE) there is no obligation to estimate precision for this parameter