13 Research methods, statistics, critical review and evidence-based practice13.13 Network analysis
13.13.13 Assortativity, connectivity and resilience
2 min read1 figure
Written by Resident Doctor Committee · Published
New in the 2027 syllabus
The College added this topic for exams from January 2027.
Who connects to whom
Assortativity asks whether nodes tend to link to nodes that resemble them.
Assortativity is the tendency for nodes with similar attributes or degrees to connect to each other. A node's degree is the number of connections it has.
Disassortativity is the opposite: unlike nodes connect preferentially. For example, high-degree nodes link to low-degree nodes.
Degree assortativity looks only at how many connections the nodes at each end of an edge have. It is not about whether the nodes share a diagnostic label or any other attribute.
Pattern | What it suggests |
|---|---|
Positive degree assortativity | Highly connected nodes tend to link to other highly connected nodes |
Negative degree assortativity | Highly connected nodes tend to link to sparsely connected nodes |
Near-zero assortativity | Little association, as measured by the chosen coefficient |
Attribute assortativity looks at shared characteristics rather than degree. When similar people are linked, this may simply reflect a shared setting or more opportunity for contact. It does not show that being similar causes people to form ties.
Choose a mixing measure that fits the type of attribute: categorical, numerical or degree-based.
Connectivity
An undirected network is connected when there is a path between every pair of nodes. A disconnected network falls into two or more separate parts, called components.
Strong connectivity applies to a directed network. It needs a directed path in both directions between every pair of nodes.
Weak connectivity ignores direction. You treat the directed edges as undirected and check whether the nodes are still joined.
A bridge is an edge whose removal increases the number of connected components.
An articulation point is a node whose removal, together with its edges, increases the number of components.
Resilience
Network resilience describes how connectivity or function changes when nodes or edges are removed or disrupted.
Random failures can have different effects from targeted removal of nodes that are highly connected or strategically placed.
Redundant paths, meaning alternative routes, can keep a network connected after one connection fails. A network that relies on a single bridge is easier to split.
You can summarise disruption by the size of the largest component and by changes in path lengths. Whether these measures matter depends on what the network represents.
Estimated symptom networks show statistical relationships between symptoms. Deleting a symptom node from the drawing does not show what treating that symptom would do clinically.