Practice Quiz

Test your understanding of Social Network Analysis concepts. Each question provides immediate feedback and explanations.

Score: 0 / 12

Centrality Questions

Q1. Which node has the highest degree in this network?
  • Center
  • A
  • B
  • C
Show Hint

Count the edges attached to each node.

Q2. In this graph, which node is the bridge?
  • A
  • D
  • E
  • H
Show Hint

Look for the edge that connects the two dense groups.

Q3. Which node likely has the highest betweenness centrality?
  • A
  • B
  • D
  • G
Show Hint

Think about which nodes all cross-community paths must pass through.

Q4. In a chain A-B-C-D-E-F, which node has the highest closeness centrality?
  • A
  • C
  • D
  • F
Show Hint

Closeness rewards being near the centre of the network.

Q5. Why can a node have low degree but high betweenness?
  • It has many self-loops
  • It's the only path between two groups
  • It has weighted edges
  • It's in the largest community
Show Hint

Think about bridge nodes connecting different parts of a network.

Q6. Why might eigenvector centrality rank a node higher than degree centrality would?
  • It has more self-loops
  • It's connected to high-degree nodes
  • It has fewer connections
  • It's isolated
Show Hint

Eigenvector centrality considers not just quantity but quality of connections.

Community Questions

Q7. Which partition looks more like natural communities?
  • A) {1,2,3}, {4,5,6,7}, {8,9,10}
  • B) {1,4,8}, {2,5,9}, {3,6,7,10}
  • C) All nodes in one group
  • D) Each node alone
Show Hint

Look for groups of nodes that are densely connected internally with few connections between groups.

Q8. What happens to modularity when a bridge node is moved to the other community?
  • Q always increases
  • Q always decreases
  • It depends on the specific network
  • Q stays the same
Show Hint

Consider both the edges gained and lost, plus the expected edge contributions.

Q9. Why is 'all nodes in one community' usually not a useful partition?
  • Q = 0 because Σ(Aij - ki·kj/2m) sums to zero over all pairs
  • It gives Q = 1
  • It's too complex
  • The formula doesn't work
Show Hint

Think about what the null model means — the expected edges were designed to sum to the actual number.

Q10. What is the role of expected edges (ki·kj/2m) in modularity?
  • They make all edge weights equal
  • They represent what we'd see in a random graph with the same degrees
  • They count triangles
  • They measure path length
Show Hint

Think about the null model — what would a random network look like?

Q11. What is the difference between overlapping and disjoint communities?
  • Overlapping communities are always larger
  • In overlapping communities, nodes can belong to multiple groups
  • Disjoint communities have no edges
  • They're the same thing
Show Hint

Think about whether a person can belong to multiple social groups at once.

Q12. What does the Louvain algorithm try to maximise?
  • Number of communities
  • Number of edges
  • Modularity Q
  • Node degree
Show Hint

Louvain is a community detection algorithm that optimises a specific metric.

Back to Home