The Unique employees across two segments setting controls the minimum difference in number of survey responses between two different groups of employees, where there is a large overlap in employees between the two groups.
This is used to ensure answers from small groups cannot be inferred through the difference between the scores for the groups, by process of deduction.
For this reason, you can view this setting as a counterpart to the Number of responses setting. While the Number of responses hides segments that are too small to view, this setting hides segments that are too large instead.
To learn more about confidentiality on Peakon, see Data visibility and confidentiality resources.
This article will contain:
As with the other data visibility settings, they can be managed by system administrators and will affect all company data.
To set the unique employees across two segments:
- Go to Administration > Data settings
- Choose the Data tab
- Set the Unique employees across two segments
Viewing overlapping segments
When the setting is enabled, each time a user is trying to access a specific segment, the system asks itself: "Does this user have access to a conflicting segment?" and if the answer is yes, it will hide one of the segments in the conflict.
When encountering a segment that is hidden due to this setting, the system will suggest an alternative segment to view. Click on See overlapping segment to proceed.
How Peakon detects conflicting segments
The Unique employees across two segments setting checks are performed in three ways:
- When filtering data by a sub-segment
- When viewing direct reports segment
- When viewing any segment with rights to view other segments
When filtering data by a sub-segment
The first check relates to a sub-segment having so many responses that it is too close to the overall context. This could be a risk as answers can potentially be inferred by comparing the segment to the context.
Setting the Unique employees across two segments setting protects the smaller segment from being exposed by hiding the big segment.
Here is an example of the kind of problem this check solves:
|A manager of 10 reports is trying to segment their data by gender. The responses are from 9 females and 1 male. The Number of responses is 3.||Segments will be visible if they have more than 3 responses. In this scenario it will be possible to infer the scores from the male employee.||
Setting the Unique employees across two segments to 2 means that segments in this context will be visible if they have between 3 and 8 responses. Both gender segments are hidden to protect the smaller segment.
Here is a similar example, illustrating that this can affect any size segments:
There are 96 employees answering the survey through email, and 4 employees using kiosk mode to answer their survey.
The company wants to compare email and kiosk based employee results.
The Number of responsesis 5.
Segments in this context will be visible if they more than 5 responses.
However, if all 96 email based employees answer, the 4 kiosk based employees may be exposed.
Setting the Unique employees across two segments to 5 would resolve this. As a result, segments in this context will be visible if they have between 5 and 95 responses.
Therefore it will not be possible to view data by kiosk or email employees separately.
Finally, here is an example where the setting would not have impact:
|A company of 116 people is looking to segment data by gender. There are responses from 57 males and 59 females. The Number of responses is 5 and the Unique employees across two segments is 3.||There is no problem, because it is not possible to infer one or the other gender segments responses. The Unique employees across two segments setting does not have impact here.||
Segments in the company context will be visible if they have between 5 and 113 responses. Therefore it will be possible to split data by both gender segments without risk.
When viewing direct reports segment
This check applies to a manager switching between their All Reports and Direct Reports segments. When the difference in responses between these two segments is below the chosen Unique employees across two segments, the direct reports segment will not be visible.
In such cases, the smallest segment is hidden to provide the most possible data:
|Marguerite Ryan is a people manager with two segments. In her segment of direct reports there are 10 responses and her all reports segment has 12 responses. Number of responses is 3. Unique employees across two segments is off.||
Both segments are visible. It could be a risk for the two people that are not in the Direct Reports segments.
|Set the Unique employees across two segments to 3. Now only the All Reports segment is visible. The Direct Reports segment is hidden to protect the two employees that do not overlap.|
When viewing any segment with rights to view other segments
The final check applies similarly to other segments in the context switcher. For example, a manager has access to their All Reports, and they also manage the Sales department segment. Therefore they can switch their context between the All Reports and Sales department segments.
If the two segments contain the same employees, with the exception of a few employees, using this setting will ensure the smallest segment does not display any survey data.
Marguerite manages 10 people through her reporting line, and also has been made into the manager of the segment Sales department that contains the same 10 all reports, herself and one additional person who reports to another manager.
Number of responses is 3. Unique employees across two segments is off.
|Both contexts are visible. Seeing as the Sales department consists of her All Reports segment + Marguerite herself + one other employee, it will be possible for her to pinpoint that person's responses.||Set the Unique employees across two segments to 3. Now only one of the segments will show data.|