Protect Relational Data Privacy
Treat every nomination as information about at least two people, minimize disclosure paths, and design governance before collecting a graph.
Method sources
By the end of this tutorial
- 1Define relational privacy and network-data governance from an explicit node, tie, boundary, direction, weight, and observation-window specification.
- 2Apply and audit this procedure: Create a data-protection impact map, minimize variables and precision, separate identifiers, encrypt and restrict raw access, define retention and deletion, test structural re-identification, aggregate outputs, and document incident response.
- 3Interpret the result with a sensitivity check and the following evidence boundary: Removing names does not anonymize a distinctive graph, and one participant cannot fully consent on behalf of every person exposed through a relational nomination.
SNA
Network specification
Analysis scenario
A school consortium plans an advice-network dashboard, but a single unusual bridge or isolate could reveal a teacher even after names are removed.
Nodes
Eligible staff whose own attributes and structural positions may be disclosed by their ties or by other people's nominations.
Ties
Sensitive directed advice nominations that contain information about both nominator and nominee, including people who did not themselves respond.
Network type
Identifiable directed whole-network data with high re-identification risk from topology, small groups, attributes, and repeated releases.
Step-by-step tutorial
Freeze the relational question
Write the decision the analysis must inform, then lock the eligible node roster, tie-generating event, direction, weight, self-tie rule, observation window, and missing-data code. Preserve a read-only source copy and record why this specification represents the stated question.
Checkpoint
A second analyst can reconstruct the same node set and edge table from the written rules without guessing what an absent record means.
Compute the focal structure
Work on a versioned analysis copy and carry out the focal method exactly as specified: Create a data-protection impact map, minimize variables and precision, separate identifiers, encrypt and restrict raw access, define retention and deletion, test structural re-identification, aggregate outputs, and document incident response. Save software and package versions, every threshold, normalization, seed, and intermediate count needed to reproduce the result.
Checkpoint
The output is tied to one named data version and includes the denominator, parameter settings, and a reproducible calculation record.
Run a structural sensitivity check
Repeat the analysis under at least one defensible alternative boundary, missingness rule, tie threshold, weight transformation, or model setting. Compare membership and substantive conclusions, not only a single coefficient, and investigate every change large enough to alter a decision.
Checkpoint
The audit states which patterns persist, which actors or groups change classification, and which conclusion depends on an analyst choice.
Report for responsible action
Pair the numerical result with a table or structure-preserving visual, document excluded and missing actors, and explain uncertainty in plain language. Convert the finding into a reversible support question, not an automatic ranking, while stating this boundary: Removing names does not anonymize a distinctive graph, and one participant cannot fully consent on behalf of every person exposed through a relational nomination.
Checkpoint
The final note contains the question, specification, result, sensitivity evidence, uncertainty, privacy controls, and a proportionate next step.
Interpret with care
relational privacy and network-data governance describes a property of the specified relation and network boundary. It does not transfer automatically to another relation, time period, class, platform, or population.
Removing names does not anonymize a distinctive graph, and one participant cannot fully consent on behalf of every person exposed through a relational nomination. Compare the result with raw counts, missingness, plausible alternative specifications, and contextual evidence before acting.