Segregation within school classes: Detecting social clustering in choice data.

Area of Science:

• Sociology
• Economics
• Education

Background:

• Social clustering and peer influence are key factors in educational and life choices.
• Distinguishing peer influence from background factors is methodologically challenging.

Purpose of the Study:

• To introduce a novel method for detecting social clustering patterns using choice data.
• To investigate social clustering among students based on shared backgrounds, considering peer influence and exogenous factors.

Main Methods:

• Developed a method comparing similar subjects within and between cohorts to isolate peer influence.
• Applied the method to Norwegian register data, analyzing student curricular track choices.

Main Results:

• Common background significantly correlates with identical curricular track choices.
• Both exogenous preferences and peer influence play a role in student choices.
• Social clustering based on shared background is observed in immigrant students and their descendants, but not in students from culturally similar countries.
• Minor effects of parental education and income on social clustering were also noted.

Conclusions:

• The developed method effectively identifies social clustering patterns driven by peer influence and background.
• Student social clustering is a complex phenomenon influenced by a combination of shared background, peer effects, and individual preferences.
• Findings highlight the importance of considering both individual and group dynamics in educational choices, particularly for immigrant populations.