RDP 2021-06: What Is Driving Participation and Diversity Trends in Economics? A Survey of High School Students Appendix B: Selection Model Results

Tanya Livermore and Mike Major

June 2021

Table B1: Likelihood of Studying Economics – Selection Models
  Coefficients
(2)   (3)   (5)
Selection
EconSchoolj		 Outcome
EconSharej		 Selection
EconSchoolj		 Outcome
EconStudentij		 Selection
EconSchoolj		 Outcome
EconStudentij
Male       0.06 0.46***   0.08 0.31**
Bilingual       0.10 0.15   0.07 0.07
Interest             −0.12* 0.27***
Understanding             0.10 0.37***
ATAR             0.11* 0.17
Non-government −0.07 −3.42***   −0.24 −0.72**   0.10 −0.46
Regional −0.10 −1.62   −0.35 0.59   −0.28 0.34
ICSEA (/100) 0.96*** 8.01***   1.20* 0.64   0.93 0.39
All boys school 1.20*** 6.51***   7.00*** −0.51   8.27*** −0.85
All girls school 0.12 −2.53***   −9.75*** −0.36   −11.87*** −0.95***
Subjects taught 0.03*** 0.04            
Subjects taught (ordinal)       0.85** −0.25   0.79* −0.31*
Year 12 cohort size (/100) 0.67***              
Year 12 cohort size (dummies)                
Fewer than 27
students		       −0.43     −0.68  
61–91 students       −0.84     −1.03  
92–134 students       3.39**     4.60***  
135 or more students       12.87***     15.93***  
Constant −11.59*** −76.69***            
Observations 768   1,995   1,141
Of which selected 316   1,238   719
Year(s) 12   11 & 12   11 & 12
Method Heckman   Heckprobit   Heckprobit
ρ     −0.40   −0.60
Wald χ2 31.32***   0.46   1.57
School-clustered standard errors     Yes   Yes
Log likelihood −1,255.49***   −863.48   −481.11

Notes: Wald χ2 is the chi-squared from a Wald test of independence of the outcome and selection equations ( H0 : ρ = 0 ); Log likelihood is the pseudolikelihood; *,** and *** denotes statistical significance at the 10, 5 and 1 per cent levels, respectively

Sources: Australian Curriculum, Assessment and Reporting Authority; NSW Education Standards Authority; RBA