ECC3841 Chap.5 Homophily and Segregation
Homophily and Segregation
Homophily is disproportionate linking among similar nodes relative to an opportunity baseline. Raw same-type counts can mislead when group sizes differ, so compare actual links with possible pairs or population composition. Choice homophily reflects preferences conditional on opportunity; induced homophily reflects schools, geography, occupations and other settings that shape who meets.
Consequences depend on what flows through ties: trust and enforcement can strengthen while novel information and outside opportunity decline.
What this chapter covers
- 01
Same-type and cross-type link counts
- 02
Opportunity denominators
- 03
Basic and adjusted homophily
- 04
Choice versus induced homophily
- 05
Group-specific mixing rates
- 06
Segregation and redundant information
- 07
Trust, sanctions and support
- 08
Policies that alter meeting opportunities
Worked example · free
Compare opportunity-adjusted link rates
- +1Amber opportunities are six, giving four divided by six.
- +1Violet opportunities are three, giving two divided by three.
- +1Cross opportunities are twelve, giving five divided by twelve.
- +1Both within-type rates exceed the cross rate.
- +1Describe homophily without claiming preference as the cause.
Key terms
- Homophily
- Excess same-type linking relative to a stated baseline.
- Choice homophily
- Similarity produced by preferences among available contacts.
- Induced homophily
- Similarity produced by the composition of meeting opportunities.
- Mixing matrix
- A table of links between each pair of node types.
- Segregation
- Separation of groups across network relationships or components.
Homophily and Segregation FAQ
Why are raw link counts insufficient?
Large groups have more possible same-type pairs. Opportunity-adjusted rates distinguish composition from unusually frequent same-type linking. Use opportunity-adjusted rates and identify the resource transmitted before drawing a segregation or welfare conclusion.
Does positive homophily prove prejudice?
No. Preferences are one mechanism, but geography, schools, occupations and institutional assignment can create the same pattern. Use opportunity-adjusted rates and identify the resource transmitted before drawing a segregation or welfare conclusion.
Can homophily have benefits?
Dense similar ties can strengthen trust, enforcement and mutual support. The same redundancy can restrict novel information and outside opportunities. Use opportunity-adjusted rates and identify the resource transmitted before drawing a segregation or welfare conclusion.
How can policy separate mechanisms?
Change meeting opportunities or exploit random exposure, then compare linking among people facing similar choice sets. Use opportunity-adjusted rates and identify the resource transmitted before drawing a segregation or welfare conclusion.
Exam move
Practice within-type n(n-1)/2 and cross-type nAmB denominators until the opportunity set is automatic. Put actual links above possible pairs before simplifying a rate. A large same-type count is not informative without the denominator. Construct a mixing table with types on rows and columns, then decide whether directed nominations or mutual ties are being counted.
Report group-specific rates because a pooled number can hide different experiences for majority and minority members. For any adjusted homophily index, write the observed same-type share, the baseline opportunity share, the scaling denominator and the sign. Keep the underlying counts beside the index, especially when a group is small or the baseline is close to one. Separate description from mechanism.
Positive adjusted mixing can come from preference, geography, schools, occupations or another opportunity structure. A credible explanation needs comparable choice sets or variation that changes meetings independently of preferences. Build pairs of examples in which the same preference operates under different group compositions and in which the same composition operates under different preferences.
This prevents the baseline from disappearing in prose. When discussing consequences, name the resource transmitted through the tie. Similar relationships may strengthen trust, sanctions and informal insurance while reducing novel job information or cross-group diffusion. Decide whether redundancy or credibility dominates for that outcome.
In policy questions, distinguish creating contact from creating durable ties and from changing downstream outcomes. Measure all three stages. During final review, calculate rates on a small two-type graph, interpret a positive and a negative adjusted value, and explain one design that separates choice from induced homophily. Never turn the index into a moral or causal label.
Working through Homophily and Segregation in ECC3841? Sia is AskSia’s AI Economics tutor — ask any ECC3841 Homophily and Segregation question and get a clear, step-by-step explanation grounded in how ECC3841 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.
