Introduction Case

Assume we have a group of 10 business engineering students working on a group project. Some of them follow data analytics (DA) and the others operations research (OR). They all belong to one of the two groups: DA or OR. These students sometimes collaborate when they are working on a group project. They can therefore be linked together to form a network. In this case, two students are connected if they are currently working together. As a result we have three types of edges in the network:

• edges that connect two DA students, denoted by ‘dd’
• edges that connect two OR students, denoted by ‘oo’
• edges that connect a DA and OR student, denoted by ‘do’.

4. Heterophilicity

Heterophilicity is defined as the ratio between the actual number of cross label edges , m_do, and the expected number of such edges in a random net.

bar_m_do <- n_da*n_or*p
H <- m_do/bar_m_do
H
[1] 0.3947368

Practice

Assume we have a group of 6 students: Tom, Sofie, Arno, Jan, Karen and Laura. Assume that Tom, Sofie and Arno are business engineering (B) students, while Jan, Karen and Laura are studying economics (E). They all belong to one of the two groups: B or E. All 6 students are going on an exchange programme next semester. Therefore, the students can be linked together based on the destination that they are going to:

As a result we have three types of edges in the network:

• edges that connect two B students, denoted by ‘bb’
• edges that connect two E students, denoted by ‘ee’
• edges that connect a B and E student, denoted by ‘be’.

Assume that:

• The number of nodes that represent business engineering students and economics students are already calculated and stored as n_b and n_e, respectively.
• The number of edges are already calculated and stored as m_bb, m_ee, and m_be.
• The connectedness of the network is already calculated and stored as p.
• The igraph library has been loaded.