...
- First, the programme computes the extrapolated number of holdings (W) which contribute to Y (i.e. which have a non-zero record value for x ) and rounds W to 0 decimals.
- Then, if that extrapolated number of holdings is higher than 0 and lower than or equal to 4 a certain value V, the programme suppresses:
- the extrapolated number of holdings (W) for cell c (if planned to be disseminated) and
- the extrapolated aggregated value of variable x(Y) for cell c .
...
- the extrapolated number of holdings in cells is rounded to the nearest multiple of 10, and
- the extrapolated aggregated values of all variables in cells are rounded to the nearest multiple of 10.
Overall assessment of the procedure and possible improvements
...
- .
...
Table 26 – Problems, possible improvements and proposals for application of primary confidentiality
...
Problem description
...
Possible improvement
...
Analysis Proposal
...
The procedure suppresses data when the number of holdings in the population is less than or equal to 4 compared to other domains where the suppression is applied when that number is less than or equal to 3. When the extrapolated number of holdings is in the interval (3; 4], the data is overprotected, as the risk of a holding (knowing its contribution and the total of a cell) to derive the other individual contributions is not realistic, except maybe if that holding is part of an enterprise operating more holdings.
...
No longer suppress data when the extrapolated number of holdings is in the interval (3; 4].
...
This improvement can be easily applied.
...
Data rounding causes inconsistencies (sums do not add up to totals). The reason is that rounding is applied to individual cells and totals in independent way. Totals on rows or columns are not calculated as the sum of the cells concerned.
...
To render the totals consistent with the sums of cells, a possible solution is the implementation of controlled rounding (using Tau-Argus). This involves rounding the tabular data to a pre-specified base while ensuring additivity of totals.
...
The controlled rounding procedure causes loss of accuracy in individual cells, by trying to maintain accuracy of totals. It might therefore be more appropriate to instead limit to warn users that cells do not add up to totals because of cells' data rounding (and suppression).
...
Knowing its own contribution and the total of a cell, the second largest contributor can estimate the minimum and maximum value of the first larger contributor. The minimum value of the first largest contributor is the value of the second largest contributor while the maximum value of the first largest contributor is the difference between the total and the second largest contributor.
...
A solution is applying the p% rule, according to which a cell is safe if the cell total minus the two largest contributors exceeds p% of the largest. This rule gives sufficient uncertainty that the second largest contributor cannot determine the size of the largest contributor.
...
It has to be assessed whether this rule provides some value added (if any), considering the very small likelihood of the risk of this disclosure in our domain, but also considering the dominance rule already in place (which is a concentration rule as the p% rule).
...
A suppressed cell can be recalculated with some margin by the difference between the total and the sum of the other cells.
...
A solution is applying secondary confidentiality. Secondary confidentiality is treating a non-confidential cell as confidential, to prevent disclosure of a confidential cell, by making it impossible for a user to recalculate the values of confidential cells.
...
Limitations to the application of secondary confidentiality
...